SYSTEME UND VERFAHREN ZUM PLANEN UND AKTUALISIEREN EINER FAHRZEUGTRAJEKTORIE

Techniken zum Generieren einer Fahrtrajektorie für ein Fahrzeug, während das Fahrzeug durch ein Objekt (z. B. ein anderes Fahrzeug, ein Fahrrad oder einen Fußgänger) blockiert wird (z. B. haben die Sensoren des Fahrzeugs ein Objekt detektiert, durch das das Fahrzeug nicht mehr in der Lage ist, sich zu bewegen), und Ausführen der Fahrtrajektorie, sobald das Fahrzeug nicht mehr blockiert wird, sind bereitgestellt. Zudem sind Techniken zum Aktualisieren eines Teils einer Fahrtrajektorie eines Fahrzeugs basierend auf einer Bestimmung, dass ein Objekt ein Segment der aktuellen Fahrtrajektorie zu einem späteren Zeitpunkt überqueren wird, ohne Neuberechnen der gesamten Trajektorie bereitgestellt.

Verfahren zum autonomen Fahren eines Fahrzeugs in einer Engstelle

Motion predition

Systems and methods for planning and updating a vehicle's trajectory

A method comprising the steps of generating operational commands for a vehicle, the operational commands being associated with segments 1710, 1725, 1730 of a trajectory 1705 of a vehicle; detecting locations and velocities of detected objects; determining that an object and the vehicle are projected to be proximate to a particular segment of the vehicle trajectory at a particular time; identifying a portion of the operational commands associated with the particular segment and updating the identified portion of the operational commands; and executing the operational commands such that the vehicle follows the updated trajectory.

ESTIMATION DEVICE, ESTIMATION METHOD, AND STORAGE MEDIUM

An estimation device includes a recognition unit configured to recognize a surrounding environment of a moving object in recognition regions, and an estimation unit configured to estimate a risk for the moving object on the basis of a recognition result from the recognition unit, in which the recognition unit sets a priority region on which a recognition process is preferentially performed among the recognition regions, according to a state of the surrounding environment of the moving object, and sets, as the priority region, a region overlapping a region including at least a part of at least one crosswalk that is present in a vicinity of an intersection region in which a first road on which the moving object is located intersects a second road present in an advancing direction of the moving object.

INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, COMPUTER PROGRAM PRODUCT, AND VEHICLE CONTROL SYSTEM

According to an embodiment, an information processing device includes one or more processors. The processors are configured to: acquire a plurality of pieces of detection information including detection results at two-dimensional positions different from each other acquired by detection of an object by one or more detection devices through a transmission body, the plurality of pieces of detection information including distortion due to the transmission body that exists between the detection devices and the object; detect a feature point from each of the plurality of pieces of detection information; and estimate, by minimizing an error between a three-dimensional position corresponding to the feature point and a detection position of the feature point corrected based on a distortion map expressing distortion at each of the two-dimensional positions, the distortion map, the three-dimensional position, and the detection position.

VEHICLE TRAVELING CONTROL METHOD, ELECTRONIC DEVICE, STORAGE MEDIUM, CHIP AND VEHICLE

The disclosure relates to the technical field of vehicle engineering, in particular to a vehicle traveling control method, an electronic device, a storage medium, a chip and a vehicle. The vehicle traveling control method includes: determining (S11) predicted motion features of a to-be-avoided object within a perception visual field of a vehicle; determining (S12) a safety level of the to-be-avoided object relative to the vehicle according to the predicted motion features and preset safety conditions; and determining (S13) a target deceleration of the vehicle according to an attribute feature of the to-be-avoided object and the corresponding safety level, and controlling traveling of the vehicle according to the target deceleration.

СПОСОБ И СИСТЕМА ДЛЯ ВЫЧИСЛЕНИЯ ДАННЫХ ДЛЯ УПРАВЛЕНИЯ РАБОТОЙ БЕСПИЛОТНОГО АВТОМОБИЛЯ

Изобретение относится к способу и системе для вычисления данных для управления работой беспилотного автомобиля. Беспилотный автомобиль (SDC) движется по сегменту дороги, который имеет полосу движения. Способ содержит этапы, на которых получают предсказанную траекторию объекта, ассоциированную с объектом в сегменте дороги, получают набор привязочных точек вдоль полосы движения. Предсказанная траектория объекта основана на данных перемещения объекта, а набор привязочных точек представляет путь транспортного средства по умолчанию для SDC вдоль полосы движения. Одна из набора привязочных точек указывает потенциальную будущую позицию SDC вдоль пути транспортного средства по умолчанию. Для каждой одной из набора привязочных точек определяют последовательность будущих моментов времени, когда SDC потенциально находится в соответствующей будущей позиции, соответствующей одной из набора привязочных точек, за счет этого формируя матричную структуру, включающую в себя будущие позиционно-временные пары ...

СПОСОБ ЗАДАНИЯ МАРШРУТА И УСТРОЙСТВО ЗАДАНИЯ МАРШРУТА

FIELD: route setting devices. SUBSTANCE: invention relates to a method and a device for setting a route. Method of setting a route using a sensor of peripheral vehicles installed in vehicle to detect positions of other vehicles moving near vehicle, and controller for setting route of vehicle according to movement paths of other vehicles based on previous positions of other vehicles. Method comprises the steps of calculating an offset value, setting the vehicle route according to the motion paths of the vehicle ahead when the offset value is less than the first threshold value, and setting the vehicle route according to the motion paths of the other vehicle, different from the vehicle ahead, when the offset value is the first threshold value or more. Offset value indicates deviation to the right and left of the vehicles moving ahead. EFFECT: higher safety of vehicle driving is achieved. 6 cl, 10 dwg РОССИЙСКАЯ ФЕДЕРАЦИЯ (19) RU (11) (13) 2 719 117 C1 (51) МПК B60W 30/165 (2012.01) ФЕДЕРАЛЬНАЯ СЛУЖБА ПО ИНТЕЛЛЕКТУАЛЬНОЙ СОБСТВЕННОСТИ (12) ОПИСАНИЕ ИЗОБРЕТЕНИЯ К ПАТЕНТУ (52) СПК B60W 30/165 (2020.02) (21)(22) Заявка: 2019112736, 26.09.2016 (24) Дата начала отсчета срока действия патента: (73) Патентообладатель(и): НИССАН МОТОР КО., ЛТД. (JP) Дата регистрации: 17.04.2020 Приоритет(ы): (22) Дата подачи заявки: 26.09.2016 (45) Опубликовано: 17.04.2020 Бюл. № 11 (85) Дата начала рассмотрения заявки PCT на национальной фазе: 26.04.2019 (86) Заявка PCT: JP 2016/078297 (26.09.2016) 2 7 1 9 1 1 7 (56) Список документов, цитированных в отчете о поиске: US 2016/0257342 A1, 08.09.2016. US 2015/0100228 A1, 09.04.2015. RU 2597066 C2, 10.09.2016. R U 26.09.2016 (72) Автор(ы): УЕДА, Хиротоси (JP) 2 7 1 9 1 1 7 R U WO 2018/055773 (29.03.2018) Адрес для переписки: 129090, Москва, ул. Б. Спасская, 25, стр. 3, ООО "Юридическая фирма Городисский и Партнеры" (54) СПОСОБ ЗАДАНИЯ МАРШРУТА И УСТРОЙСТВО ЗАДАНИЯ МАРШРУТА (57) Реферат: Изобретение относится к способу и устройству едущего впереди ...

Fahrzeugpositionsverarbeitungsvorrichtung, Fahrzeugsteuerungsvorrichtung, Fahrzeugpositionsverarbeitungsverfahren und Fahrzeugsteuerungsverfahren

Es wird bereitgestellt eine Fahrzeugpositionsverarbeitungsvorrichtung, eine Fahrzeugsteuerungsvorrichtung, ein Fahrzeugpositionsverarbeitungsverfahren und ein Fahrzeugsteuerungsverfahren, die in der Lage sind, die Anzahl der zur Erzeugung der Trajektorie verwendeten Positionsinformationen des vorderen Objekts zu erhöhen und die Erzeugungsgenauigkeit der Trajektorie zu verbessern. Fahrzeugpositionsverarbeitungsvorrichtung, eine Fahrzeugsteuerungsvorrichtung, ein Fahrzeugpositionsverarbeitungsverfahren und ein Fahrzeugsteuerungsverfahren, das Positionen eines Zielobjekts erhält, einen Trajektorienerzeugungsbereich einstellt, der ein kontinuierlicher Bereich ist, der eine Position des Zielobjekts nahe einer Position des vorliegenden eigenen Fahrzeugs enthält, Positionen des Zielobjekts, die in dem Trajektorienerzeugungsbereich enthalten sind, unter den mehreren Positionen des Zielobjekts als Zielobjektpositionen für die Trajektorienerzeugung auswählt und eine Trajektorie des Zielobjekts basierend ...

Object tracking supporting autonomous vehicle navigation

Trajectory planning of vehicles using route information

An autonomous vehicle receives information from vehicle sensors indicating the presence of an object (e.g. bus, delivery vehicle, bicycle) nearby. The object is identified (e.g. a bus with a particular model number) and the expected route of the object is retrieved (e.g. from a database, vehicle-to-vehicle communications or a mobile device associated with the object). The current trajectory is estimated using the on-board sensors and is compared with the expected route of the object, including approximating where the object may be in 5 seconds. The information is used to plan the trajectory of the autonomous vehicle (e.g. to avoid collisions). The expected route may be a one-time or temporary route (e.g. recreational drivers, delivery vehicles, construction vehicles) or a re-occurring route (e.g. postal services, busses). A confidence level may be assigned based on an accuracy and/or reliability of the expected route of the object.

Predicting the behavior of a vehicle using agent-to-agent relations to control an autonomous vehicle

Method of predicting the trajectory 5’ of a pedestrian 5, and based on the predicted pedestrian behaviour 5’, an action (arrow) of vehicle 2b can be predicted. For example, if the pedestrian 5 is predicted to enter the vehicle 2b, the vehicle 2b can be predicted to pull away soon. Depending on the predicted trajectory of the vehicle 2b, the host vehicle 1 may prepare to perform evasive action. Multiple pedestrian trajectories may be predicted. Time series of state vector may be generated from camera data for predicting future trajectory 5’ of the pedestrian 5. A graph representing a relationship between the pedestrian and the parked vehicle 2b may be generated. Based on the predicted outcome, the host vehicle 1 may determine a driving trajectory or a parking area.

Navigation of autonomous vehicles using turn aware machine learning based models for prediction of behavior of a traffic entity

An autonomous vehicle collects sensor data of an environment surrounding the autonomous vehicle including traffic entities such as pedestrians, bicyclists, or other vehicles. The sensor data is provided to a machine learning based model along with an expected turn direction of the autonomous vehicle to determine a hidden context attribute of a traffic entity given the expected turn direction of the autonomous vehicle. The hidden context attribute of the traffic entity represents factors that affect the behavior of the traffic entity, and the hidden context attribute is used to predict future behavior of the traffic entity. Instructions to control the autonomous vehicle are generated based on the hidden context attribute.

Perception and motion prediction for autonomous devices

Systems, methods, tangible non-transitory computer-readable media, and devices associated with object perception and prediction of object motion are provided. For example, a plurality of temporal instance representations can be generated. Each temporal instance representation can be associated with differences in the appearance and motion of objects over past time intervals. Past paths and candidate paths of a set of objects can be determined based on the temporal instance representations and current detections of objects. Predicted paths of the set of objects using a machine-learned model trained that uses the past paths and candidate paths to determine the predicted paths. Past path data that includes information associated with the predicted paths can be generated for each object of the set of objects respectively.

MOTOR VEHICLE DRIVER ASSISTANCE METHOD

СПОСОБЫ И ПРОЦЕССОРЫ ДЛЯ УПРАВЛЕНИЯ РАБОТОЙ БЕСПИЛОТНОГО АВТОМОБИЛЯ

FIELD: transport industry.SUBSTANCE: invention relates to self-driving cars. Disclosed is a method for controlling the operation of a self-driving car (SDC). The method is carried out by an electronic device connected to the SDC. In this case, the method includes the stages at which, at the first moment of time during the approach of the SDC to the pedestrian crossing, the zone near the pedestrian crossing is identified by means of an electronic device, the presence of the object in the zone including the objects is determined, the time interval for the object is determined based on the movement data of the object. The method also includes the steps of using the time interval and the SDC movement data to determine data for operation control, to control the operation of the SDC, and assign the decision data indicating the decision to the zone including the objects in such a way so that the priority of the path between the SDC and any other object identified by another unique identifier that is different from the first unique identifier in the object-containing region is the same as between the SDC and the object.EFFECT: timely planning of SDC operation is achieved when approaching a pedestrian crossing.14 cl, 7 dwg РОССИЙСКАЯ ФЕДЕРАЦИЯ (19) RU (11) (13) 2 750 118 C1 (51) МПК B60W 30/00 (2006.01) B60K 31/00 (2006.01) G01C 21/20 (2006.01) G01C 21/26 (2006.01) G01C 21/34 (2006.01) G08G 1/16 (2006.01) ФЕДЕРАЛЬНАЯ СЛУЖБА ПО ИНТЕЛЛЕКТУАЛЬНОЙ СОБСТВЕННОСТИ (12) ОПИСАНИЕ ИЗОБРЕТЕНИЯ К ПАТЕНТУ (52) СПК B60W 30/00 (2021.02); B60K 31/00 (2021.02); G01C 21/20 (2021.02); G01C 21/26 (2021.02); G01C 21/34 (2021.02); G08G 1/16 (2021.02) (21)(22) Заявка: 2019143920, 25.12.2019 25.12.2019 (73) Патентообладатель(и): Общество с ограниченной ответственностью "Яндекс Беспилотные Технологии" (RU) Дата регистрации: 22.06.2021 (45) Опубликовано: 22.06.2021 Бюл. № 18 2 7 5 0 1 1 8 R U (54) СПОСОБЫ И ПРОЦЕССОРЫ ДЛЯ УПРАВЛЕНИЯ РАБОТОЙ БЕСПИЛОТНОГО АВТОМОБИЛЯ (57) Реферат: Изобретение относится ...

ВАРИАНТЫ УПРАВЛЕНИЯ РАБОТОЙ АВТОНОМНОГО ТРАНСПОРТНОГО СРЕДСТВА

FIELD: autonomous vehicle management and autonomous driving. SUBSTANCE: invention relates to the management of an autonomous vehicle and autonomous driving. Driving an autonomous vehicle through the vehicle transport network includes actuation of a performance control evaluation module specimen for a specific version of the operation; accepting the performance of the candidate vehicle control from the performance control evaluation module specimen for a specific version of the operation, and driving through parts of the vehicle transport network in accordance with the candidate action of the vehicle operation. Therein performance control evaluation module specimen for a specific version of the operation includes a copy of the performance control evaluation model for a particular version of the vehicle’s operation. A vehicle operation version is a vehicle operation version in merging with or a vehicle operation version of bypassing an obstacle. EFFECT: an autonomous vehicle’s operational safety is increased. 15 cl, 9 dwg РОССИЙСКАЯ ФЕДЕРАЦИЯ (19) RU (11) (13) 2 744 640 C1 (51) МПК G01C 21/26 (2006.01) B60W 30/08 (2012.01) ФЕДЕРАЛЬНАЯ СЛУЖБА ПО ИНТЕЛЛЕКТУАЛЬНОЙ СОБСТВЕННОСТИ (12) ОПИСАНИЕ ИЗОБРЕТЕНИЯ К ПАТЕНТУ (52) СПК G01C 21/26 (2020.08); B60W 30/08 (2020.08) (21)(22) Заявка: 2020120980, 30.11.2017 (24) Дата начала отсчета срока действия патента: Дата регистрации: Приоритет(ы): (22) Дата подачи заявки: 30.11.2017 (56) Список документов, цитированных в отчете о поиске: US 9534910 B2, 03.01.2017. JP 2016017914 A, 01.02.2016. RU 2459259 C1, 20.08.2012. (45) Опубликовано: 12.03.2021 Бюл. № 8 (85) Дата начала рассмотрения заявки PCT на национальной фазе: 30.06.2020 (86) Заявка PCT: US 2017/064089 (30.11.2017) 2 7 4 4 6 4 0 (73) Патентообладатель(и): НИССАН НОРТ АМЕРИКА, ИНК. (US), ДЗЕ ЮНИВЕРСИТИ ОФ МАССАЧУСЕТС (US) 12.03.2021 R U 30.11.2017 (72) Автор(ы): РЭЙ, Кайл Холлинз (US), ВИТВИКИ, Стефан (US), ЗИЛЬБЕРШТЕЙН, Шломо (US) 2 7 4 4 6 4 0 R U WO 2019/108213 (06.06.2019) ...

Verfahren zum Betrieb eines autonom fahrenden Fahrzeuges

Die Erfindung betrifft ein Verfahren zum Betrieb eines autonom fahrenden Fahrzeuges (1), wobei erfindungsgemäß vorgesehen ist, dass bei einem erfassten Fahrkorridor eines eine Kurve (K) durchfahrenden, entgegenkommenden Fahrzeuges (2), wobei der Fahrkorridor zumindest aufgrund der Abmessungen des entgegenkommenden Fahrzeuges (2) in eine Fahrspur (F1) des Fahrzeuges (1) hineinragt, das Fahrzeug (1) an einem ermittelten Haltepunkt vor der Kurve (K) in den Stillstand versetzt wird und/oder derart verzögert wird, dass dem entgegenkommenden Fahrzeug (2) ausreichend Freiraum zum Durchfahren der Kurve (K) zur Verfügung gestellt wird.

SYSTEME UND VERFAHREN ZUM NAVIGIEREN EINES FAHRZEUGS

Ein autonomes System kann die Kontrolle des menschlichen Fahrers über ein Host-Fahrzeug selektiv ablösen. Das System kann ein Bild empfangen, das repräsentativ für eine Umgebung des Host-Fahrzeugs ist, und auf der Grundlage der Analyse des Bildes ein Hindernis in der Umgebung des Host-Fahrzeugs erkennen. Das System kann eine Fahrereingabe an eine mit dem Host-Fahrzeug assoziierte Drossel-, Brems- und/oder Lenksteuerung überwachen. Das System kann bestimmen, ob die Fahrereingabe dazu führen würde, dass das Host-Fahrzeug innerhalb eines Näherungspuffers relativ zum Hindernis navigiert. Wenn die Fahrereingabe nicht dazu führen würde, dass das Host-Fahrzeug innerhalb des Näherungspuffers navigiert, kann das System der Fahrereingabe erlauben, eine korrespondierende Änderung in einem oder mehreren Host-Fahrzeug-Bewegungssteuerungssystemen zu verursachen. Wenn die Fahrereingabe dazu führen würde, dass das Host-Fahrzeug innerhalb des Näherungspuffers navigiert, kann das System verhindern, dass ...

SYSTEM AND METHOD FOR AVOIDING CONTACT BETWEEN AUTONOMOUS AND MANNED VEHICLES CAUSED BY LOSS OF TRACTION

A control system for preventing vehicle collisions may include a vehicle location determination module (212), a terrain determination module (222), a terrain surface coefficient of friction estimation module (224), and a sensing system configured to generate signals indicative of vehicle speed, vehicle pose, vehicle size, vehicle weight, vehicle tire type, vehicle load, vehicle gear ratio, weather characteristics, and road conditions for a vehicle operating at a job site. A manned vehicle trajectory determination module (210) may receive location information and plot a first travel path (142) for a manned vehicle (140) based at least in part on a location, heading, and speed of the manned vehicle (140) and a desired destination for the manned vehicle (140). An autonomous vehicle trajectory determination module (220) may receive location information, terrain information, and terrain surface coefficient of friction information, plot a second travel path (122) for an autonomous vehicle (120 ...

TRAJECTORY PLANNER

An autonomous vehicle capable of trajectory prediction may include a first sensor, a second sensor, a processor, a trajectory planner, a low-level controller, and vehicle actuators. The first sensor may be of a first sensor type and may detect an obstacle and a goal. The second sensor may be of a second sensor type and may detect the obstacle and the goal. The processor may perform matching on the obstacle detected by the first sensor and the obstacle detected by the second sensor, model an existence probability of the obstacle based on the matching, and track the obstacle based on the existence probability and a constant velocity model. The trajectory planner may generate a trajectory for the autonomous vehicle based on the tracked obstacle, the goal, and a non-linear model predictive control (NMPC). The low-level controller may implement the trajectory for the autonomous vehicle by driving vehicle actuators.

Method and apparatus for autonomous driving control, electronic device, and storage medium

The present application discloses a method and an apparatus for autonomous driving control, an electronic device, and a storage medium; the application relates to the technical field of autonomous driving. A specific implementation solution is: obtaining movement data of a pedestrian, where the movement data includes a velocity component of the pedestrian along a width direction of a lane and a time of duration that the pedestrian cuts into a driving path of the autonomous vehicle from one side; determining a movement direction of the pedestrian according to the movement data and the movement information of the pedestrian; and generating a driving strategy for the autonomous vehicle according to the movement direction of the pedestrian. Therefore, the movement direction of the pedestrian can be accurately predicted, which facilitates the autonomous vehicle to avoid the pedestrian and insures driving safety.

SYSTEM, METHOD, AND COMPUTER PROGRAM PRODUCT FOR TRAJECTORY SCORING DURING AN AUTONOMOUS DRIVING OPERATION IMPLEMENTED WITH CONSTRAINT INDEPENDENT MARGINS TO ACTORS IN THE ROADWAY

Provided are autonomous vehicles (AV), computer program products, and methods for maneuvering an AV in a roadway, including receiving forecast information associated with predicted trajectories of one or more actors in a roadway, determining a relevant trajectory of an actor based on correlating a forecast for predicted trajectories of the actor with the trajectory of the AV, regenerate a distance table for the relevant trajectory previously generated for processing constraints, generate a plurality of margins for the AV to evaluate, the margins based on a plurality of margin types for providing information about risks and effects on passenger comfort associated with a future proximity of the AV to the actor, classifying an interaction between the AV and the actor based on a plurality of margins, and generating continuous scores for each candidate trajectory that is also within the margin of the actor generated for the relevant trajectory.

FAHRZEUGZIELVERFOLGUNG

Die Offenbarung stellt ein Zielverfolgungssystem bereit. Ein System umfasst einen Computer, der einen Prozessor und einen Speicher beinhaltet, wobei der Speicher Anweisungen speichert, die vom Prozessor ausführbar sind, um ein Ziel in Lidar-Punktwolkendaten zu verfolgen, indem eine Fehlerfunktion auf Grundlage einer geglätteten Zielposition, einer geglätteten Zielgeschwindigkeit, einer geglätteten Zielbeschleunigung und einer gemessenen Zielposition minimiert wird. Der Prozessor kann ferner programmiert sein, um ein Fahrzeug auf Grundlage der Verfolgung des Ziels zu betreiben.

Verfahren zum Erstellen einer probabilistischen Freiraumkarte mit statischen und dynamischen Objekten

Die Erfindung betrifft ein Verfahren zum Erstellen einer probabilistischen Freiraumkarte mit statischen (2a, 2b, 3) und dynamischen Objekten (V1-V7) mit den folgenden Schritten:- Abrufen (S1) statischer Objekte (2a, 2b, 3) sowie eines Wahrnehmungsbereichspolygons (WP) aus einem vorhandenen Umfeldmodell;- Sammeln (S2) von prädizierten Trajektorien (T1, T2) von dynamischen Objekten (V1-V7);- Zusammenfügen (S3) der statischen Objekte (2a, 2b, 3), des Wahrnehmungsbereichspolygons (WP) und der prädizierten Trajektorien (T1, T2) in einer ersten Freiraumkarte;- Festlegen (S4) einer maximalen Prädiktionszeit;- Festlegen (S5) von Prädiktionszeitschritten;- Festlegen (S6) einer aktuellen Prädiktionszeit und setzen dieser aktuellen Prädiktionszeit auf den Wert 0 zum Festlegen des Starts eines festgelegten Prädiktionszeitraums;- Festlegen (S7) von Konfidenzbereichen (K) um die statischen (2a, 2b, 3) und dynamischen Objekte (V1-V7);- Festlegen (S8) zumindest eines unsicheren Bereichs (U) um zumindest ...

Navigating multi-way stop intersections with an autonomous vehicle

A system comprising one or more computer processors; and one or more non-transitory storage media storing instructions which, when executed by the one or more computer processors, cause performance of operations comprising: while a first vehicle is operating in an autonomous mode at a multi-way stop intersection and has a highest precedence at the multi-way stop intersection: detecting, using a processing circuit, movement of a second vehicle at the intersection, the second vehicle (1302e) having an expected travel path (1314) through the intersection that intersects a planned travel path (1312) of the first vehicle (100) through the intersection; in accordance with a determination, based on the detected movement of the second vehicle, that the second vehicle is expected to exit the intersection, instructing, using a control circuit, the first vehicle to proceed into the intersection before the second vehicle exits the intersection. Trajectory and speed may be used to determine the movement ...

Positioning at least one vehicle in relation to a set of moving targets

A method and system for positioning a vehicle in relation to each moving target of an ordered set of moving targets. Each of the moving targets moves from an initial position at a constant velocity. Embodiments can compute (602) an estimated time for the vehicle to be positioned within a predetermined proximity of one of the moving targets; compute (604) an estimated location of the moving target at the estimated time, based on a current position of the moving target and the constant velocity of the moving target, and compute (606) a required velocity for the vehicle to move from its current position to reach the estimated location by the estimated time. If the required velocity is less than or equal to a maximum velocity of the vehicle, outputting (312) the estimated time and the estimated location for use in positioning the vehicle.

Target arrangement, method, and control unit for following a tar-get vehicle

Method (400), control unit (230), and target arrangement (100) of a leading vehicle (101), for triggering a follower vehicle (102), which is situated at a lateral distance from the leading vehicle (101), to coordinate its movements with the leading vehicle (101). The target arrangement (100) comprises a target (110), configured to be placed at a lateral distance from to the leading vehicle (101). The target (110) is also configured to be recognised by at least one forwardly (105) directed sensor (130) of the follower vehicle (102).

SYSTEM AND METHOD FOR AVOIDING CONTACT BETWEEN AUTONOMOUS AND MANNED VEHICLES CAUSED BY LOSS OF TRACTION

A control system for preventing vehicle collisions may include a vehicle location determination module, a terrain determination module, a terrain surface coefficient of friction estimation module, and a sensing system configured to generate signals indicative of vehicle speed, vehicle pose, vehicle size, vehicle weight, vehicle tire type, vehicle load, vehicle gear ratio, weather characteristics, and road conditions for a vehicle operating at a job site. A manned vehicle trajectory determination module may receive location information and plot a first travel path for a manned vehicle based at least in part on a location, heading, and speed of the manned vehicle and a desired destination for the manned vehicle. An autonomous vehicle trajectory determination module may receive location information, terrain information, and terrain surface coefficient of friction information, plot a second travel path for an autonomous vehicle, and determine projected slide trajectories for the autonomous ...

SYSTEMS AND METHODS FOR NAVIGATING A VEHICLE

A system for navigating a host vehicle may: receive, from an image capture device, an image representative of an environment of the host vehicle; determine a navigational action for accomplishing a navigational goal of the host vehicle; analyze the image to identify a target vehicle in the environment of the host vehicle; determine a next-state distance between the host vehicle and the target vehicle that would result if the navigational action was taken; determine a maximum braking capability of the host vehicle, a maximum acceleration capability of the host vehicle, and a speed of the host vehicle; determine a stopping distance for the host vehicle; determine a speed of the target vehicle and assume a maximum braking capability of the target vehicle; and implement the navigational action if the stopping distance for the host vehicle is less than the next-state distance summed together with a target vehicle travel distance.

Methods and systems for computer-based determining of presence of dynamic objects

A method for determining a set of dynamic objects in sensor data representative of a surrounding area of a vehicle having sensors, the method being executed by a server, the server executing a machine learning algorithm (MLA). Sensor data is received, and the MLA generates, based on the sensor data, a set of feature vectors. Vehicle data indicative of a localization of the vehicle is received. The MLA generates, based on the set of feature vectors and the vehicle data, a tensor, the tensor including a grid representation of the surrounding area. The MLA generates an mobility mask indicative of grid cells occupied by at least one moving potential object in the grid, and a velocity mask indicative of a velocity associated with the at least one potential object in the grid. The MLA determines, based on the mobility mask and the velocity mask, the set of dynamic objects.

Target arrangement, method, and control unit for following a target vehicle

Method, control unit, and target arrangement of a leading vehicle for triggering a follower vehicle, which is situated at a lateral distance from the leading vehicle, to coordinate its movements with the leading vehicle. The target arrangement comprises a target configured to be placed at a lateral distance from to the leading vehicle. The target is also configured to be recognized by at least one forwardly directed sensor of the follower vehicle.

Systems and methods for estimating cuboids from LiDAR, map and image data

Systems and methods for operating a robotic system. The methods comprise: inferring, by a computing device, a first heading distribution for the object from a 3D point cloud; obtaining, by the computing device, a second heading distribution from a vector map; obtaining, by the computing device, a posterior distribution of a heading using the first and second heading distributions; defining, by the computing device, a cuboid on a 3D graph using the posterior distribution; and using the cuboid to facilitate driving-related operations of a robotic system.

Road User Information Determination Based on Image and Lidar Data

A method of determining information related to a road user in an environment of a vehicle includes receiving, from vehicle sensors, a digital image and a Lidar point cloud. The digital image and the Lidar point cloud represent a scene in the environment of the vehicle. The method includes detecting a road user in the scene based on the received digital image and Lidar point cloud. The method includes generating a combined digital representation of the detected road user by combining corresponding image data and Lidar data associated with the detected road user. The method includes determining information related to the detected road user by processing the combined digital representation of the detected road user.

METHOD FOR CONTROLLING AN AUTONOMOUS VEHICLE FOR ALL TYPES OF SCENARIO

Procédé de contrôle automatique d'un premier véhicule (V1) automobile, caractérisé en ce qu'il comprend : - une étape (E1) de détection d'un ensemble d'obstacles mobiles (V2, V3, V4) environnant le premier véhicule, - une étape (E2) de définition d'une voie de circulation virtuelle (VCV), puis - une étape (E3) de définition d'obstacles virtuels (W2, W3, W4) par une projection des obstacles mobiles dans la voie de circulation virtuelle, - une étape (E4) de détermination de positions futures des obstacles virtuels dans la voie de circulation virtuelle, - une étape (E5) de détermination de délais d'interaction des obstacles virtuels avec le premier véhicule, - une étape (E6) de contrôle du premier véhicule en fonction des délais d'interaction des obstacles virtuels avec le premier véhicule.

LANE HANDLING FOR MERGE PRIOR TO TURN

Ortsvorhersage für dynamische Objekte

Die vorliegende Erfindung betrifft ein Steuerungssystem und ein Verfahren zur Vorhersage eines Orts von dynamischen Objekten, beispielsweise von Fußgängern, die von den Sensoren eines Fahrzeugs erfasst werden können. Das Steuerungssystem (10) weist eine Vielzahl von Sensoren (30) auf und ein Rechensystem (40), das eingerichtet ist, die Objekte (20), welche von der Vielzahl von Sensoren (30) erfasst sind, mittels eines ersten Programms zu einer Objektliste (22) zu kombinieren, wobei jeder Eintrag der Objektliste (22) für jedes der Objekte (20) den Ort, eine Geschwindigkeit und eine freie Strecke umfasst, und die Objektliste (22) einen Zeitstempel enthält; und aus einer vordefinierten Anzahl von Objektlisten (22), mittels eines zweiten Programms, für mindestens einen Teil der dynamischen Objekte (20) eine weitere Objektliste (23) zu bestimmen, wobei die weitere Objektliste (23) einen Zeitstempel für einen zukünftigen Zeitpunkt enthält und mindestens den Ort der dynamischen Objekte (20) umfasst ...

AUTONOMOUS VEHICLE OPERATIONAL MANAGEMENT SCENARIOS

Traversing, by an autonomous vehicle, a vehicle transportation network, may include operating a scenario-specific operational control evaluation module instance, wherein the scenario-specific operational control evaluation module instance includes an instance of a scenario-specific operational control evaluation model of a vehicle operational scenario wherein the vehicle operational scenario is a merge vehicle operational scenario or a pass-obstruction vehicle operational scenario, receiving a candidate vehicle control action from the scenario-specific operational control evaluation module instance, and traversing a portion of the vehicle transportation network in accordance with the candidate vehicle control action.

AUTONOMOUS VEHICLE OPERATIONAL MANAGEMENT SCENARIOS

Traversing, by an autonomous vehicle, a vehicle transportation network, may include operating a scenario-specific operational control evaluation module instance, wherein the scenario-specific operational control evaluation module instance includes an instance of a scenario-specific operational control evaluation model of a vehicle operational scenario wherein the vehicle operational scenario is a merge vehicle operational scenario or a pass-obstruction vehicle operational scenario, receiving a candidate vehicle control action from the scenario-specific operational control evaluation module instance, and traversing a portion of the vehicle transportation network in accordance with the candidate vehicle control action.

ROUTE SETTING METHOD AND ROUTE SETTING DEVICE

Provided is a route setting method with which a vehicle can stably travel so as to continuously follow a travel path of another vehicle such as a preceding vehicle. A route setting method in which are used a peripheral vehicle sensor that is installed in a host vehicle and that detects the position of another vehicle traveling in the periphery of the host vehicle, and a controller that sets the route of the host vehicle on the basis of a travel path determined from a history of the positions of the other vehicle, wherein: the amount of change in the travel path of a preceding vehicle among other vehicles is calculated (S6); and when the amount of change in the travel path of the preceding vehicle is equal to or greater than a threshold value, the route of the host vehicle is set on the basis of the travel path of another vehicle different from the preceding vehicle (S7-S10).

Secure autonomous conduit in the event of detection of a target object

System and method for vehicle occlusion detection

A system and method for vehicle occlusion detection is disclosed. A particular embodiment includes: receiving training image data from a training image data collection system; obtaining ground truth data corresponding to the training image data; performing a training phase to train a plurality of classifiers, a first classifier being trained for processing static images of the training image data, a second classifier being trained for processing image sequences of the training image data; receiving image data from an image data collection system associated with an autonomous vehicle; and performing an operational phase including performing feature extraction on the image data, determining a presence of an extracted feature instance in multiple image frames of the image data by tracing the extracted feature instance back to a previous plurality of N frames relative to a current frame, applying the first trained classifier to the extracted feature instance if the extracted feature instance ...

PROACTIVE SAFE DRIVING FOR AN AUTOMATED VEHICLE

A method and corresponding apparatus involve monitoring, by a first motor vehicle, a position of a second motor vehicle in an adjacent lane and performing an automated safety routine. The safety routine includes determining, based on sensor data, whether the vehicles are maintaining a same speed and determining whether the first motor vehicle can switch to the adjacent lane without colliding with the second motor vehicle. If the vehicles are maintaining the same speed and the first motor vehicle cannot switch to the adjacent lane, a longitudinal offset is set based on the sensor data and established by automatically decreasing the speed of the first motor vehicle. The speed of the first motor vehicle is automatically adjusted to maintain at least the longitudinal offset until the first motor vehicle can switch to the adjacent lane without colliding with the second motor vehicle.

SYSTEMS AND METHODS FOR ANTICIPATORY LANE CHANGE

An anticipatory lane change system is described for assisting a host vehicle positioned in a current lane that is adjacent an adjacent lane. The anticipatory lane change system may include an identification module that identifies a potential lane change location and receives proximate vehicle data associated with proximate vehicles. The anticipatory lane change system may include a prediction module that predicts future kinematic data at a future time for a set of the proximate vehicles. The anticipatory lane change system may include a determination module that determines whether a gap will be available at the potential lane change location at the future time based on the future kinematic data. The anticipatory lane change system may include a lane change module that initiates a lane change maneuver for the host vehicle in response to determining that the gap will be available at the potential lane change location at the future time.

SYSTEMS AND METHODS FOR NAVIGATING A VEHICLE

Systems and methods are provided for vehicle navigation. In one implementation, a processing device may be configured to obtain a planned driving action for accomplishing a navigational goal of a host vehicle; receive sensor data from an environment surrounding the host vehicle; identify a target vehicle moving in the environment and a velocity of the target vehicle; calculate, a predicted trajectory for the target vehicle; calculate a planned trajectory for the host vehicle corresponding to the planned driving action; identify an intersection of the planned trajectory for the host vehicle with the predicted trajectory for the target vehicle; determine a braking action of the host vehicle to comply with a safety requirement; and cause the braking action to be applied to decelerate the host vehicle to change the planned trajectory of the host vehicle, until the planned trajectory does not intersect the predicted trajectory of the target vehicle.

OCCULSION AWARE PLANNING AND CONTROL

Techniques are discussed for controlling a vehicle, such as an autonomous vehicle, based on occluded areas in an environment. An occluded area can represent areas where sensors of the vehicle are unable to sense portions of the environment due to obstruction by another object. An occlusion grid representing the occluded area can be stored as map data or can be dynamically generated. An occlusion grid can include occlusion fields, which represent discrete two- or three-dimensional areas of driveable environment. An occlusion field can indicate an occlusion state and an occupancy state, determined using LIDAR data and/or image data captured by the vehicle. An occupancy state of an occlusion field can be determined by ray casting LIDAR data or by projecting an occlusion field into segmented image data. The vehicle can be controlled to traverse the environment when a sufficient portion of the occlusion grid is visible and unoccupied.

MOBILE-BODY INFORMATION ACQUIRING SYSTEM, MOBILE-BODY INFORMATION ACQUIRING METHOD, PROGRAM, AND MOBILE BODY

The purpose of the present invention is to provide a mobile-body information acquiring system, a mobile-body information acquiring method, a program, and a mobile body, with which accuracy of information acquired by a mobile body serving as a subject can be improved. Provided is a mobile-body information acquiring system wherein: a calculation unit has a processing unit for setting a time point or a time when a first mobile body - desires to take action about movement or a time point or a time in the very near future that the first mobile body - desires to know in relation to movement; and a communication unit transmits information related to a time point or a time including time point information expressed by absolute time points. 1. A mobile-body information acquiring system that exchanges information between mobile bodies including a movement unit for movement , a calculation unit that performs calculation relating to the movement , and a communication unit that moves in combination with the movement unit and performs communication of an output from the calculation unit as information ,wherein the calculation unit includes a processing unit that sets a time point or time at which a first mobile body desires to take an action about movement or a time point or time in the very near future which the first mobile body desires to know in relation to movement, andthe communication unit transmits information of the time point or the time which includes time point information expressed by an absolute time point.2. The mobile-body information acquiring system according to claim 1 ,wherein the time point or the time that is transmitted by the communication unit is a time point after predetermined time from a current time point or a transmission time point.3. The mobile-body information acquiring system according to claim 1 ,wherein the time point or the time that is set by the processing unit includes a plurality of time points or a plurality of kinds of time in the future ...

SENSOR FUSION FOR AUTONOMOUS MACHINE APPLICATIONS USING MACHINE LEARNING

In various examples, a multi-sensor fusion machine learning model – such as a deep neural network (DNN) – may be deployed to fuse data from a plurality of individual machine learning models. As such, the multi-sensor fusion network may use outputs from a plurality of machine learning models as input to generate a fused output that represents data from fields of view or sensory fields of each of the sensors supplying the machine learning models, while accounting for learned associations between boundary or overlap regions of the various fields of view of the source sensors. In this way, the fused output may be less likely to include duplicate, inaccurate, or noisy data with respect to objects or features in the environment, as the fusion network may be trained to account for multiple instances of a same object appearing in different input representations.

Scenario identification for validation and training of machine learning based models for autonomous vehicles

A system uses a machine learning based model to determine attributes describing states of mind and behavior of traffic entities in video frames captured by an autonomous vehicle. The system classifies video frames according to traffic scenarios depicted, where each scenario is associated with a filter based on vehicle attributes, traffic attributes, and road attributes. The system identifies a set of video frames associated with ground truth scenarios for validating the accuracy of the machine learning based model and predicts attributes of traffic entities in the video frames. The system analyzes video frames captured after the set of video frames to determine actual attributes of the traffic entities. Based on a comparison of the predicted attributes and actual attributes, the system determines a likelihood of the machine learning based model making accurate predictions and uses the likelihood to generate a navigation action table for controlling the autonomous vehicle.

СПОСОБ И СИСТЕМА ДЛЯ ВЫЧИСЛЕНИЯ ДАННЫХ ДЛЯ УПРАВЛЕНИЯ РАБОТОЙ БЕСПИЛОТНОГО АВТОМОБИЛЯ

РОССИЙСКАЯ ФЕДЕРАЦИЯ (19) RU (11) (13) 2019 143 947 A (51) МПК B60W 30/00 (2006.01) G05D 1/00 (2006.01) ФЕДЕРАЛЬНАЯ СЛУЖБА ПО ИНТЕЛЛЕКТУАЛЬНОЙ СОБСТВЕННОСТИ (12) ЗАЯВКА НА ИЗОБРЕТЕНИЕ (21)(22) Заявка: 2019143947, 25.12.2019 (71) Заявитель(и): Общество с ограниченной ответственностью "Яндекс Беспилотные технологии" (RU) Приоритет(ы): (22) Дата подачи заявки: 25.12.2019 (43) Дата публикации заявки: 25.06.2021 Бюл. № 18 (72) Автор(ы): Федоров Сергей Дмитриевич (RU) Стр.: 1 A 2 0 1 9 1 4 3 9 4 7 R U A (57) Формула изобретения 1. Компьютерно-реализуемый способ формирования управляющих данных для управления работой беспилотного автомобиля (SDC), причем SDC движется по сегменту дороги, причем сегмент дороги имеет полосу движения, причем способ осуществляется посредством электронного устройства, соединенного с SDC, при этом способ содержит этапы, на которых- получают, посредством электронного устройства, предсказанную траекторию объекта, ассоциированную с объектом в сегменте дороги, причем предсказанная траектория объекта основана на данных перемещения объекта; получают, посредством электронного устройства, набор привязочных точек вдоль полосы движения, причем набор привязочных точек представляет путь транспортного средства по умолчанию для SDC вдоль полосы движения, причем данная одна из набора привязочных точек указывает потенциальную будущую позицию SDC вдоль пути транспортного средства по умолчанию; для каждой одной из набора привязочных точек определяют, посредством электронного устройства, последовательность будущих моментов времени, когда SDC потенциально находится в соответствующей будущей позиции соответствующей одной из набора привязочных точек, за счет этого формируя матричную структуру, включающую в себя будущие позиционно-временные пары для SDC, причем данная будущая позиционно-временная пара указывает то, когда и где SDC должен быть потенциально расположен в будущем вдоль пути транспортного средства по умолчанию; для каждой будущей позиционно-временной пары в ...

Navigating multi-way stop intersections with an autonomous vehicle

A system, method and storage media for operating an autonomous vehicle (AV) at a multi-way stop intersection. After detecting the AV is at a primary stopline (1306) of the multi-way stop intersection, a planned travel path though the multiway stop intersection is obtained. If the planned travel path of the AV through the multi-way stop intersection satisfies a set of one or more clearance criteria, the AV proceeds past the primary stopline (1306). The clearance criteria include a criterion that is satisfied in response to detecting the AV is clear to safely merge into a travel lane (1312) corresponding to the planned travel path. The system may also use a secondary stopline (1316) if the path ahead of the vehicle is obstructed or it is not safe to pass this point immediately. System may also use speed and distance thresholds to determine safe entry to the travel lane.

Systems and methods for planning and updating a vehicle's trajectory

A method of controlling a vehicle comprising the steps of: receiving sensor data representing objects in an environment of the vehicle; determining that the vehicle is blocked by one of the objects; determining probable locations for objects based on the sensor data and a timestamp of the sensor data; generating one or more operational commands for the vehicle while the vehicle is blocked, based on the probable locations of objects; and executing the one or more operational commands, comprising manoeuvring the vehicle along a path that is unblocked by the object. Also provided is a computer-readable storage medium, and a vehicle.

Tracker position updates for vehicle trajectory generation

A method is disclosed for updating tracker position when generating a trajectory for a vehicle may include receiving, from a detection and tracking system of a vehicle, a first position of an object at a first time. A first trajectory of the object may be determined based on at least on the first position of the object at the first time. A second portion of the object at a second time may be received from the detection and tracking system. A second trajectory for the object may be generated to include an initial waypoint corresponding to the second position of the object at the second time, and a final waypoint corresponding to a final waypoint of the first trajectory.

Motion planning and intention prediction for autonomous driving in highway scenarios via graphical model-based factorization

Aspects of the disclosure are related to a method, apparatus, and system for planning a motion for a first vehicle, comprising: estimating past states of an observed second vehicle based on sensor inputs; predicting a future trajectory of the second vehicle based on the estimated past states; planning a future trajectory of the first vehicle based on the predicted future trajectory of the second vehicle and a safety cost function; and driving the first vehicle to follow the planned trajectory.

All Mover Priors

Systems, devices, products, apparatuses, and/or methods for generating a driving path for an autonomous vehicle on a roadway by determining one or more prior probability distributions of one or more motion paths for one or more objects that have previously moved in a geographic location and/or for controlling travel of an autonomous vehicle on a roadway by predicting movement of a detected object according to one or more prior probability distributions of one or more motion paths for one or more objects that have previously moved in a geographic location.

Depth dependent pixel filtering

Sensors, including time-of-flight sensors, may be used to detect objects in an environment. In an example, a vehicle may include a time-of-flight sensor that images objects around the vehicle, e.g., so the vehicle can navigate relative to the objects. Sensor data generated by the time-of-flight sensor can include returns associated with highly reflective objects that cause glare. In some examples, a depth of a sensed surface is determined from the sensor data and additional pixels at the same depth are identified. The subset of pixels at the depth are filtered by comparing a measured intensity value to a threshold intensity value for the depth. Other threshold intensity values can be applied to subsets of pixels at different depths.

OBJEKTVERFOLGUNG ZUR UNTERSTÜTZUNG AUTONOMER FAHRZEUGNAVIGATION

Diese Offenbarung betrifft allgemein Systeme und Verfahren zum optischen Verfolgen von Objekten in der Nähe eines autonomen Fahrzeugs. Insbesondere betrifft sie ein System zur Objektverfolgung, das Positionsdaten für verfolgte Objekte verfeinern kann, indem es einen Standort der Objekte in der Umgebung des autonomen Fahrzeugs zumindest teilweise auf der Grundlage zuvor ermittelter Standorte der Objekte bestimmt. In bestimmten Fällen können die prädizierten und detektierten Standorte, die verwendet werden, um zu einem verfeinerten Standort für die Objekte zu gelangen, in Abhängigkeit von den Bedingungen der Sensordaten und der Qualität der historischen Daten unterschiedlich gewichtet werden.

Systems and methods for planning and updating a vehicle's trajectory

DEVICE, METHOD, AND STORAGE MEDIUM

A device includes a storage device configured to store a program; and a hardware processor, wherein, the hardware processor executes the program stored in the storage device to: recognize positions of a plurality of traffic participants; determine a temporary goal which each of the plurality of traffic participants is trying to reach in the future, based on the recognition results; and simulate a movement process in which each of the plurality of traffic participants moves toward the temporary goal using a movement model to estimate an action in the future of each of the plurality of traffic participants.

Transportation Network for Multi-featured Autonomous Vehicles

A system for operation of an autonomous transportation network and a method of operation for a plurality of multi-featured autonomous vehicles are disclosed. The system comprises a road, a control management center and a plurality of multi-featured autonomous vehicles. The multi-featured autonomous vehicles include different types of vehicles for transportation of passengers or goods in the autonomous transportation network. The method of operation disclosed comprises selecting permissible routes from an origin to a destination for the multi-featured autonomous vehicles and predicting conflicts for the multi-featured autonomous vehicles. Conflict avoidance instructions are generated and are transmitted to the multi-featured autonomous vehicles using infrastructure elements. The method of operation comprises adjusting the route of the multi-featured autonomous vehicles (20) using corrected route instructions calculated by an onboard processor of the multi-featured autonomous vehicles.

Full uncertainty for motion planning in autonomous vehicles

Systems and methods for motion planning by a vehicle computing system of an autonomous vehicle are provided. The vehicle computing system can input sensor data to a machine-learned system including one or more machine-learned models. The computing system can obtain, as an output of the machine-learned model(s), motion prediction(s) associated with object(s) detected by the system. The system can convert a shape of the object(s) into a probability of occupancy by convolving an occupied area of the object(s) with a continuous uncertainty associated with the object(s). The system can determine a probability of future occupancy of a plurality of locations in the environment at future times based at least in part on the motion prediction(s) and the probability of occupancy of the object(s). The system can provide the motion prediction(s) and the probability of future occupancy of the plurality of locations to a motion planning system of the autonomous vehicle.

Bird's eye view based velocity estimation

Systems and methods determining velocity of an object associated with a three-dimensional (3D) scene may include: a LIDAR system generating two sets of 3D point cloud data of the scene from two consecutive point cloud sweeps; a pillar feature network encoding data of the point cloud data to extract two-dimensional (2D) bird's-eye-view embeddings for each of the point cloud data sets in the form of pseudo images, wherein the 2D bird's-eye-view embeddings for a first of the two point cloud data sets comprises pillar features for the first point cloud data set and the 2D bird's-eye-view embeddings for a second of the two point cloud data sets comprises pillar features for the second point cloud data set; and a feature pyramid network encoding the pillar features and performing a 2D optical flow estimation to estimate the velocity of the object.

SYSTEMS AND METHODS FOR TRAJECTORY BASED SAFEKEEPING OF VEHICLES

Verfahren zum autonomen Fahren eines Fahrzeugs in einer Engstelle

Die Erfindung betrifft ein Verfahren zum autonomen Fahren eines Fahrzeugs (1) durch eine Engstelle (5). Das Verfahren umfasst ein Hinterlegen einer Vorfahrtsregelung für die Engstelle (5) in einer Datenbank, auf welche das Fahrzeug (1) zugreifen kann. Die Engstelle (5) und ein im Bereich der Engstelle (5) entgegenkommendes Gegenfahrzeug (13) werden mit Sensoren des Fahrzeugs erfasst und es erfolgt ein Ermitteln der Geschwindigkeit des Gegenfahrzeugs (13) mittels von den Sensoren erfasster Geschwindigkeits-Daten. Weiterhin wird eine Reaktion des Gegenfahrzeugs (13) in Abhängigkeit von der ermittelten Geschwindigkeit des Gegenfahrzeugs (13) vorausgesagt, und das Fahrzeug (1) wird durch die Engstelle (5) verfahren, sofern das Voraussagen der Reaktion des Gegenfahrzeugs (13) ergibt, dass das Gegenfahrzeug (13) die Engstelle nicht passieren wird oder die Engstelle (5) zum Passieren freigibt oder die in der Datenbank hinterlegte Vorfahrtsregelung vorsieht, dass das Fahrzeug (1) in der Engstelle ...

Verfahren zum automatischen Queren eines Kreuzungsbereichs mit einem Fahrzeug

Die Erfindung betrifft ein Verfahren zum automatischen Queren eines Kreuzungsbereichs (1) mit einem Fahrzeug (2), bei dem durch eine Sensorprüfung von vom Fahrzeug (2) umfassten Sensoren ein Vorliegen von weiteren Verkehrsteilnehmern (3) innerhalb eines Detektionsbereichs (4) des Fahrzeugs (2) feststellbar ist, wobei im Falle eines im Detektionsbereich (4) vorliegenden weiteren Verkehrsteilnehmers (3) ein sicheres Queren des Kreuzungsbereichs (1) möglich ist, solange der weitere Verkehrsteilnehmer (3) einen minimalen Abstand (5*) zum Fahrzeug (2) noch nicht unterschritten hat, und nach Unterschreitung des minimalen Abstands (5*) das Queren des Kreuzungsbereichs (1) unterbrechbar ist. Das erfindungsgemäße Verfahren ist gekennzeichnet durch folgende Verfahrensschritte:- Initiieren eines Anfahrvorgangs zum Queren des Kreuzungsbereichs (1);- Durchführen einer Sensorprüfung, wobei eine Beendigung einer relevanten Sensorprüfung zum rechtzeitigen Detektieren eines weiteren Verkehrsteilnehmers ...

System and method for proximate vehicle intention prediction for autonomous vehicles

A system and method for proximate vehicle intention prediction for autonomous vehicles are disclosed. A particular embodiment is configured to: receive perception data associated with a host vehicle; extract features from the perception data to detect a proximate vehicle in the vicinity of the host vehicle; generate a trajectory of the detected proximate vehicle based on the perception data; generate, using a trained intention prediction model, a predicted intention of the detected proximate vehicle based on the perception data and the trajectory of the detected proximate vehicle; generate, using the predicted intention of the detected proximate vehicle, a predicted trajectory of the detected proximate vehicle; and output the predicted intention and predicted trajectory for the detected proximate vehicle to another subsystem.

POSITIONING AT LEAST ONE VEHICLE IN RELATION TO A SET OF MOVING TARGETS

A method and system for positioning a vehicle in relation to each moving target of an ordered set of moving targets. Each of the moving targets moves from an initial position at a constant velocity. Embodiments can compute (602) an estimated time for the vehicle to be positioned within a predetermined proximity of one of the moving targets; compute (604) an estimated location of the moving target at the estimated time, based on a current position of the moving target and the constant velocity of the moving target, and compute (606) a required velocity for the vehicle to move from its current position to reach the estimated location by the estimated time. If the required velocity is less than or equal to a maximum velocity of the vehicle, outputting (312) the estimated time and the estimated location for use in positioning the vehicle.

Target arrangement, method, and control unit for following atar-get vehicle

Method (400), control unit (230), and target arrangement (100) of a leading vehicle (101), for triggering a follower vehicle (102), which is situated at a lateral distance from the leading vehicle (101), to coordinate its movements with the leading vehicle (101). The target arrangement (100) comprises a target (110), configured to be placed at a lateral distance from to the leading vehicle (101). The target (110) is also configured to be recognised by at least one forwardly (105) directed sensor (130) of the follower vehicle (102).

LOCATION PREDICTION FOR DYNAMIC OBJECTS

A control system and a method for predicting a location of dynamic objects, for example, of pedestrians, which are able to be detected by the sensors of a vehicle. The control system includes a multitude of sensors and a processing system, which is configured to combine with a first program the objects that are detected by the multitude of sensors to form an object list, each entry of the object list encompassing the location, a speed and an open route for each of the objects, and the object list including a time stamp; and to determine with a second program for at least a portion of the dynamic objects an additional object list from a predefined number of object lists, the additional object list including a time stamp for a future point in time and encompassing at least the location of the dynamic objects.

System and method for controlling motion of vehicle in shared environment

A method controls a motion of the host vehicle in the environment according to a trajectory and adjusts the trajectory of the vehicle based on the levels of risk posed by the motion of other vehicles. The method determines a set of feasible trajectories of hypothetical vehicles traveling in a driving area of the host vehicle and determines a level of risk of each feasible trajectory as a combination of the probability of the feasible trajectory to intersect with the trajectory of the host vehicle and the probability of the feasible trajectory to be followed by at least one vehicle. The method adjusts the trajectory of the host vehicle in response to assessing the levels of risk of the feasible trajectories.

Verfahren und Vorrichtung zum Bereitstellen mindestens einer Trajektorie für ein automatisiert gefahrenes Fahrzeug

Die Erfindung betrifft ein Verfahren zum Bereitstellen mindestens einer Trajektorie (20) für ein automatisiert gefahrenes Fahrzeug (50), wobei ein zukünftiges Verhalten von im Umfeld des Fahrzeugs erkannten Objekten (60) mittels einer Prädiktionseinrichtung (2) prädiziert wird, wobei Prädiktionen (Px) für jedes der Objekte (60) jeweils eine Eintreffenswahrscheinlichkeit zugeordnet wird, wobei die Prädiktionen (Px) in Abhängigkeit der jeweils zugeordneten Eintreffenswahrscheinlichkeiten einer von mindestens zwei Bewertungskategorien (A,B;C) zugeordnet werden, wobei mittels einer Trajektorienplanungseinrichtung (3) aus einer Menge von Trajektorienkandidaten (40,41,42) die mindestens eine Trajektorie (20) derart ausgewählt wird, dass die ausgewählte mindestens eine Trajektorie (20) mit Prädiktionen (Px), denen jeweils die Bewertungskategorie (A,B,C) mit der höchsten Eintreffenswahrscheinlichkeit zugeordnet ist, kollisionsfrei ist und dass ausgehend von der ausgewählten mindestens einen Trajektorie ...

Navigating muti-way stop intersections with an autonomous vehicle

Secure autonomous conduit in the event of detection of a target object

L’invention concerne un procédé de conduite autonome d’un véhicule, appelé égo-véhicule, pour la prise en compte d’une détection d’un objet cible, comprenant les étapes de : Obtention (20) de données d’au moins un capteur de l’égo-véhicule, lesdites données comprenant une information de présence de l’objet cible ; Calcul (42) d’un indicateur de présence configuré pour quantifier la probabilité de présence de l’objet cible sur une trajectoire de l’égo-véhicule ; Calcul (44) d’un indicateur de cohérence configuré pour quantifier la prédictibilité du mouvement de l’objet cible ; si l’indicateur de présence est inférieur à une deuxième valeur prédéterminée, génération (48) d’une deuxième instruction de conduite configurée pour réduire la vitesse de l’égo-véhicule ; si l’indicateur de cohérence est inférieur à une première valeur prédéterminée, génération (48) d’une première instruction de conduite configurée pour augmenter la distance entre l’objet cible et l’égo-véhicule. FIG. 2 The invention relates to a method for autonomous driving of a vehicle, called ego-vehicle, for taking into account a detection of a target object, comprising the steps of: Obtaining (20) data from at least one sensor of the ego-vehicle, said data comprising target object presence information; Calculation (42) of a presence indicator configured to quantify the probability of presence of the target object on a trajectory of the ego-vehicle; Calculation (44) of a coherence indicator configured to quantify the predictability of the movement of the target object; if the presence indicator is less than a second predetermined value, generating (48) a second driving instruction configured to reduce the speed of the ego-vehicle; if the consistency indicator is less than a first predetermined value, generating (48) a first driving instruction configured to increase the distance between the target object and the ego-vehicle. FIG. 2

Safety-Aware Comparator for Redundant Subsystems in Autonomous Vehicles

A method, system and device are disclosed for determining safety conflicts in redundant subsystems of autonomous vehicles. Each redundant subsystem calculates a world model or path plan, including locations, dimensions, and orientations of moving and stationary objects, as well as projected travel paths for moving objects in the future. The travel paths and projected future world models are subsequently compared using a geometric overlay operation. If at future time moments the projected world models match within predefined margins, the comparison results in a match. In case of a mismatch at a given future moment between projected world models, a determination is made as to whether the autonomous vehicle and all road users in this future moment are safe from collision or driving off the drivable space or road based on a geometric overlay operation.

Social behavior for autonomous vehicles

Understanding the intent of human drivers and adapting to their driving styles is used to increased efficiency and safety of autonomous vehicles (AVs) by enabling them to behave in safe and predictable ways without requiring explicit inter-vehicle communication. A Social Value Orientation (SVO), which quantifies the degree of an agent's selfishness or altruism, is estimated by the AV for other vehicles to better predict how they will interact and cooperate with others. Interactions between agents are modeled as a best response game wherein each agent negotiates to maximize their own utility. A dynamic game solution uses the Nash equilibrium, yielding an online method of predicting multi-agent interactions given their SVOs. This approach allows autonomous vehicles to observe human drivers, estimate their SVOs, and generate an autonomous control policy in real time.

OBJEKTVERFOLGUNG FÜR FAHRZEUGE

Die vorliegende Offenbarung sieht eine Objektverfolgung für Fahrzeuge vor. Vorgesehen ist ein System, umfassend einen Computer, der einen Prozessor und einen Speicher beinhaltet, wobei der Speicher Anweisungen speichert, die vom Prozessor ausführbar sind, um eine Objektstandortvorhersage auf der Grundlage eines Videodatenstroms zu bestimmen, wobei die Objektstandortvorhersage auf dem Verarbeiten zugeschnittener TEDA-Daten anhand eines neuronalen Netzwerks beruht. Der Prozessor kann ferner zum Herunterladen der Objektstandortvorhersage in ein Fahrzeug auf der Grundlage eines Standorts des Fahrzeugs programmiert sein.

Systems and methods for planning and updating a vehicle's trajectory

VEHICLE DRIVING SYSTEM AND CONTROL METHOD PERFORMED BY VEHICLE DRIVING SYSTEM

A vehicle driving system enabling safe traveling on a narrow road and a method performed by the vehicle driving system for controlling a vehicle are disclosed. The vehicle driving system includes a sensor configured to monitor an environment outside a vehicle and a controller. The controller is configured to recognize an oncoming traveling object through the sensor, derive a width of a travelable road between the traveling object and the vehicle, determine priority between the traveling object and the vehicle when the width of the travelable road is smaller than a sum of widths of the traveling object and the vehicle, and, when the traveling object is determined to have the priority, control traveling of the vehicle to provide a space in a lateral direction of the vehicle to allow the traveling object to pass.

DEVICE FOR AND METHOD OF PREDICTING A TRAJECTORY FOR A VEHICLE

Steuerung eines Kraftfahrzeugs

Ein Verfahren umfasst Schritte des Erfassens einer ersten Trajektorie eines ersten Kraftfahrzeugs beim Durchfahren einer vorbestimmten Strecke; des Bestimmens eines Verlaufs der Strecke auf der Basis vorbestimmter Kartendaten; des Bestimmens einer Güte, mit der die Trajektorie und der Verlauf übereinstimmen; und des Übermittelns der Güte an ein zweites Kraftfahrzeug.An Bord eines zweiten Kraftfahrzeugs kann ein Verlauf einer Strecke auf der Basis vorbestimmter Kartendaten bestimmt werden; eine dem Verlauf zugeordnete Güte erfasst werden; eine Umgebung des zweiten Kraftfahrzeug abgetastet werden; eine zweite Trajektorie durch Fusionieren des Verlaufs mit der Abtastung bestimmt werden; und das zweite Kraftfahrzeug kann gesteuert werden, der zweiten Trajektorie zu folgen.

Trajectory planning of vehicles using route information

Object tracking supporting autonomous vehicle navigation

A system and method comprising obtaining tracking data associated with an object, the tracking data including a first position of the object at a first time; predicting 1502 a second position of the object at a second time; capturing an image that includes the object using a sensor of an autonomous vehicle at the second time; detecting 1504 the object at a third position using the image; and in accordance with determining that the third position is within a threshold distance of the second position; determining 1506 a fourth position of the object at the second time based on the predicted second position and detected third position; and navigating the autonomous vehicle in accordance with the fourth position of the object.

TARGET ARRANGEMENT, METHOD, AND CONTROL UNIT FOR FOLLOWING A TARGET VEHICLE

TASK SCHEDULING METHOD, APPARATUS, DEVICE, AND COMPUTER READABLE STORAGE MEDIUM

The present disclosure provides a task scheduling method, an apparatus, a device and a computer readable storage medium provided by the, and an implementation solution thereof includes: identifying obstacle information of an obstacle around a vehicle; determining a safety level of the obstacle according to driving information and the obstacle information of the vehicle; determining a driving task according to the obstacle information, determining a safety level of the driving task according to the safety level of the obstacle corresponding to the obstacle information; and performing a task scheduling according to the safety level of the driving task. The method, the apparatus, the device and the computer readable storage medium provided by the present disclosure can perform the task with the highest safety level preferentially and avoid a situation in which an urgent situation is unable to be dealt with in time. 1. A task scheduling method , comprising:identifying obstacle information of an obstacle around a vehicle;determining a safety level of the obstacle according to driving information and the obstacle information of the vehicle;determining a driving task according to the obstacle information, and determining a safety level of the driving task according to the safety level of the obstacle corresponding to the obstacle information; andperforming a task scheduling according to the safety level of the driving task.2. The method according to claim 1 , wherein the identifying obstacle information of an obstacle around a vehicle claim 1 , comprises:acquiring environmental data of an environment around the vehicle through a sensor provided in the vehicle; andidentifying the obstacle according to the environmental data, and determining the obstacle information of each obstacle.3. The method according to claim 2 , wherein the sensor comprises at least one of the following:a camera, a radar and a lidar.4. The method according to claim 1 , wherein the determining a safety ...

Techniques for maintaining vehicle formations

A method of maintaining vehicle formation includes receiving a desired formation distance between a lead vehicle and a follower vehicle; receiving a pre-planned path for the follower vehicle; and defining a dynamic zone around a current position of the lead vehicle. The dynamic zone has a boundary characterized by a first radius from the current position of the lead vehicle. The first radius can be substantially equal to the desired formation distance. The method further includes determining a next speed of the follower vehicle based on a current position of the follower vehicle with respect to the boundary of the dynamic zone; determining a commanded curvature of the follower vehicle based on the current position of the follower vehicle with respect to the pre-planned path; and outputting the next speed and the commanded curvature to a control system of the follower vehicle for navigation of the follower vehicle.

Auction-based cooperative perception for autonomous and semi-autonomous driving systems

This document describes techniques, apparatuses, and systems that can implement auction-based cooperative perception for autonomous and semi-autonomous driving systems. The described techniques, apparatuses, and systems cooperatively use perception systems of an autonomous or semi-autonomous vehicle (AV) in a fleet of connected AVs to provide perception data to the entire fleet. An AV of the fleet is selected to act as an auction system (e.g., an auctioneer) and sends an announcement offering perception tasks to the fleet for bidding. The AVs of the fleet determine whether they have the communication and computational capabilities to perform the tasks and, if so, submit bids to perform one or more tasks. The auctioneer awards the tasks, and the bid-winning AV(s) perform the tasks and update the fleet. In this way, the described techniques, apparatuses, and systems can provide perception services with increased coverage and quality, which can make autonomous and semi-autonomous driving systems ...

Systems and methods for planning and updating a vehicle's trajectory

Systsems and methods for planning and updating a vehicle's trajectory

Environmental limitation and sensor anomaly system and method

Method of determining sensor condition and environmental condition based on sensor measurement, and generating a perception visibility model (PVM) indicating sensor detection capabilities (e.g. occlusion map indicating where an object is located that is occluding the sensors, or a probability of detection map 905 indicating the likelihood of an objection being detected in a given location), and planning a trajectory based on PVM. Perception pipeline may generate intermediate perception data e.g. LIDAR semantic point cloud including ground plane and detected objects. System may determine if the PVM is to be updated based on sensor and environmental conditions; if not, reusing the prior PVM as the current PVM. Prior PVM may be obtained offline 910 from a database or online 915 based on past PVMs. PVM may indicate an average perception visibility of a non-occluded object with a direct line-of-sight within a range. Sensor condition may be malfunction or occluded sensors. Environmental condition ...

SYSTEM AND METHOD FOR NAVIGATION WITH EXTERNAL DISPLAY

System, methods, and other embodiments described herein relate to selecting a route for a vehicle to travel. In one embodiment, the detection system generates a driving maneuver recommendation for a vehicle having a plurality of sensors configured to acquire information about an environment around the vehicle, the sensors including at least a camera to capture one or more images of a scene within the environment, by determining that at least a portion of each image in a set of images captured by the camera indicates an external display in the environment, tracking an object within the portion of each image in the set of images to determine a state of the object, the state including at least a trajectory estimate for the object, and determining a recommended driving maneuver based at least in part on the determined state of the object.

TECHNOLOGY TO APPLY DRIVING NORMS FOR AUTOMATED VEHICLE BEHAVIOR PREDICTION

Systems, apparatuses and methods may provide for technology that generates a series of time-stamped object graphs based on object trajectory histories derived from external object data for a plurality of external objects, such as vehicles. The technology may also generate, via a first neural network such as a graph attention network, a series of relational object representations based on the series of time-stamped object graphs, and determine, via a second neural network such as a long short-term memory network, predicted object trajectories for the plurality of external objects based on the series of relational object representations. The technology may also modify behavior of an autonomous vehicle based on the predicted object trajectories and real-time perceptual error information. 1. A computing system comprising:a sensor interface to receive external object data; and generate a series of time-stamped object graphs based on object trajectory histories derived from the external object data for a plurality of external objects;', 'generate, via a first neural network, a series of relational object representations based on the series of time-stamped object graphs; and', 'determine, via a second neural network, a prediction of future object trajectories for the plurality of external objects based on the series of relational object representations., 'a processor coupled to the sensor interface, the processor including one or more substrates and logic coupled to the one or more substrates, wherein the logic is implemented at least partly in one or more of configurable logic or fixed-functionality hardware logic, the logic coupled to the one or more substrates to2. The system of claim 1 , wherein the logic coupled to the one or more substrates is further to:include real-time perceptual error information with the predicted object trajectories; andmodify behavior of an autonomous vehicle based on the predicted object trajectories and the real-time perceptual error ...

VEHICLE OBJECT TRACKING

A system, comprising a computer that includes a processor and a memory, the memory storing instructions executable by the processor to determine an object location prediction based on a video data stream, wherein the object location prediction is based on processing cropped TEDA data with a neural network. The processor can be further programmed to download the object location prediction to a vehicle based on a location of the vehicle.

DRIVING ASSISTANCE METHOD AND DRIVING ASSISTANCE DEVICE

Verfahren und Vorrichtung zum Betrieb eines Assistenzsystems eines Fahrzeuges und Fahrzeug

Die Erfindung betrifft ein Verfahren und eine Vorrichtung zum Betrieb eines Assistenzsystems eines Fahrzeuges (1), wobei ein Abstand des Fahrzeuges (1) zu seitlich statischen und seitlich dynamischen Objekten als laterale Begrenzungsobjekte (B) erfasst werden, wobei eine Geschwindigkeit der seitlich dynamischen Objekte ermittelt und zumindest die seitlich dynamischen Objekte klassifiziert werden, dadurch gekennzeichnet, dass in einer Steuereinheit des Fahrzeuges (1) eine Kennlinienschar hinterlegt ist, deren Kennlinien (K1 bis K5) jeweils einer Umgebungssituation zugeordnet ist, wobei mittels einer jeweiligen Kennlinie (K1 bis K5) vorgegeben wird, mit welcher Maximalgeschwindigkeit (vm) das Fahrzeug (1) bei unterschiedlichen Abständen zu erfassten lateralen Begrenzungsobjekten (B) an diesen vorbeifahren soll. Weiterhin betrifft die Erfindung ein Fahrzeug (1) mit einer solchen Vorrichtung.

VERFAHREN ZUM AUTONOMEN BETREIBEN EINES FAHRZEUGS, STEUERUNGSVORRICHTUNG FÜR EIN FAHRZEUG UND FAHRZEUG

Ein Verfahren zum Betreiben eines autonom fahrenden Fahrzeugs beinhaltet ein Betreiben des Fahrzeugs in einem ersten autonomen Fahrmodus mittels einer Steuerungsvorrichtung basierend auf Sensordaten, die von einem Sensorsystem des Fahrzeugs erfasst werden, ein Ermitteln aus Umgebungsdaten, die zumindest die Sensordaten beinhalten, eines Vorhandenseins einer Übergabebedingung an einer Übergabestelle innerhalb einer geplanten Trajektorie des Fahrzeugs, ein Herstellen einer Datenkommunikation mit einem Führungsfahrzeug, das in einem autonomen Fahrmodus betrieben wird, um entlang einer Führungstrajektorie zu fahren, welche die Übergabestelle beinhaltet, und ein Betreiben des Fahrzeugs in einem zweiten autonomen Fahrmodus zumindest teilweise basierend auf ersten Hilfsdaten, die von dem Führungsfahrzeug bereitgestellt werden.

MOBILER-KÖRPER-INFORMATIONSERFASSUNGSSYSTEM, MOBILER-KÖRPER-INFORMATIONSERFASSUNGSVERFAHREN, PROGRAMM UND MOBILER KÖRPER

Der Zweck der vorliegenden Erfindung ist es, ein Mobiler-Körper-Informationserfassungssystem, ein Mobiler-Körper-Informationserfassungsverfahren, ein Programm und einen mobilen Körper anzugeben, mit denen Genauigkeit von Information, die von einem als Subjekt dienenden mobilen Körper erfasst wird, verbessert werden kann. Angegeben wird ein Mobiler-Körper-Informationserfassungssystem, in dem eine Recheneinheit eine Prozesseinheit enthält, um einen Zeitpunkt oder eine Zeit, zu dem oder der ein erster mobiler Körper eine Aktion zur Bewegung vornehmen möchte, oder einen Zeitpunkt oder eine Zeit in der sehr nahen Zukunft, den oder die der erste mobiler Körper im Bezug auf Bewegung kennen möchte, zu setzen, und eine Kommunikationseinheit, die Information in Bezug auf den Zeitpunkt oder die Zeit, der oder die durch einen absoluten Zeitpunkt ausgedrückte Zeitpunktinformation enthält, sendet.

Systems and methods for planning and updating a vehicle's trajectory

Systems and methods for deriving an agent trajectory based on multiple image sources

Examples disclosed herein may involve a computing system that is operable to (i) receive a first sequence of images captured by a monocular camera associated with a vehicle during a given period of operation and a second sequence of image pairs captured by a stereo camera associated with the vehicle during the given period of operation, (ii) derive, from the first sequence of images captured by the monocular camera, a first track for a given agent that comprises a first sequence of position information for the given agent, (iii) derive, from the second sequence of image pairs captured by the stereo camera, a second track for the given agent that comprises a second sequence of position information for the given agent, and (iv) determine a trajectory for the given agent based on the first and second tracks for the given agent.

Systems and Methods for Generating Motion Forecast Data for a Plurality of Actors with Respect to an Autonomous Vehicle

A computing system can be configured to input data that describes sensor data into an object detection model and receive, as an output of the object detection model, object detection data describing features of the plurality of the actors relative to the autonomous vehicle. The computing system can generate an input sequence that describes the object detection data. The computing system can analyze the input sequence using an interaction model to produce, as an output of the interaction model, an attention embedding with respect to the plurality of actors. The computing system can be configured to input the attention embedding into a recurrent model and determine respective trajectories for the plurality of actors based on motion forecast data received as an output of the recurrent model. 1. A computing system , comprising:an object detection model configured to receive an input representation that describes sensor data, and in response to receipt of the input representation that describes the sensor data, output object detection data describing features of a plurality of actors relative to an autonomous vehicle;an interaction model configured to receive an input sequence that describes the object detection data, and in response to receipt of the input sequence, generate an attention embedding with respect to the plurality of actors;a recurrent model configured to receive the attention embedding, and in response to receipt of the attention embedding, generate motion forecast data with respect to the plurality of actors, the motion forecast data describing respective trajectories for the plurality of actors;a memory that stores a set of instructions; input the input representation that describes the sensor data into the object detection model;', 'receive, as an output of the object detection model, the object detection data describing the features of the plurality of the actors relative to the autonomous vehicle;', 'generate an input sequence that describes the ...

Robust Optimization-Based Affine Abstractions For Uncertain Affine Dynamics

A method for affine abstraction of intention models of vehicles is disclosed. 1. A method in a data processing system comprising at least one processor and at least one memory , the at least one memory comprising instructions executed by the at least one processor to implement an affine abstraction generation process for dynamics of a second vehicle , the method comprising:receiving, from a plurality of sensors coupled to an ego vehicle, second vehicle data about the second vehicle, the second vehicle data comprising a set of values associated with at least a portion of an augmented state;determining a parameter of the second vehicle based on the second vehicle data and an affine abstraction for an intention model associated with the second vehicle, the affine abstraction previously generated by minimizing an approximation error subject to a set of constraints by solving a linear problem; andproviding the parameter of the second vehicle to a vehicle control system coupled to the ego vehicle.2. The method of claim 1 , wherein the linear problem is a single level linear programming problem.3. The method of claim 1 , wherein the set of constraints is predetermined based on the augmented state.4. The method of claim 1 , wherein the affine abstraction comprises a pair of hyperplanes.5. The method of claim 4 , wherein the hyperplanes bound a domain of possible driving behaviors.6. The method of claim 1 , wherein the intention model comprises a proportional-derivative control input.7. The method of claim 1 , wherein the intention model is one of a malicious intention model or a cautious intention model.8. The method of claim 1 , wherein the parameter of the second vehicle is a future velocity of the second vehicle claim 1 , and the method further comprises:determining an intention of the second vehicle based on the velocity of the second vehicle at a future time; andproviding the intention to the vehicle control system.9. The method of claim 8 , wherein the parameter is ...

Systems and Methods for Mitigating Vehicle Pose Error Across an Aggregated Feature Map

Systems and methods for improved vehicle-to-vehicle communications are provided. A system can obtain sensor data depicting its surrounding environment and input the sensor data (or processed sensor data) to a machine-learned model to perceive its surrounding environment based on its location within the environment. The machine-learned model can generate an intermediate environmental representation that encodes features within the surrounding environment. The system can receive a number of different intermediate environmental representations and corresponding locations from various other systems, aggregate the representations based on the corresponding locations, and perceive its surrounding environment based on the aggregated representations. The system can determine relative poses between the each of the systems and an absolute pose for each system based on the representations. Each representation can be aggregated based on the relative or absolute poses of each system and weighted according to an estimated accuracy of the location corresponding to the representation. 1. A computer-implemented method , the method comprising:obtaining, by a computing system comprising one or more computing devices onboard an autonomous vehicle, sensor data associated with an environment of a first autonomous vehicle;obtaining, by the computing system, estimated location data indicative of a first estimated pose of the first autonomous vehicle;determining, by the computing system, a first intermediate environmental representation of at least a first portion of the environment of the first autonomous vehicle based, at least in part, on the sensor data;obtaining, by the computing system, a first message from a second autonomous vehicle, wherein the first message comprises a second intermediate environmental representation of at least a second portion of the environment of the first autonomous vehicle and second estimated location data indicative of a second estimated pose of the second ...

VEHICLE TRAJECTORY PLANNING METHOD AND ELECTRONIC DEVICE

A vehicle trajectory planning method and an electronic device are provided, which are related to the field of automatic driving and intelligent road network. The method includes: receiving initially planned trajectories of respective autonomous vehicles; and adjusting, in a case that a conflict exists between the initially planned trajectories of the respective autonomous vehicles, a planned trajectory of an autonomous vehicle associated with the conflict. 1. A vehicle trajectory planning method , comprising:receiving initially planned trajectories of respective autonomous vehicles; andadjusting, in a case that a conflict exists between the initially planned trajectories of the respective autonomous vehicles, a planned trajectory of an autonomous vehicle associated with the conflict.2. The vehicle trajectory planning method of claim 1 , wherein the initially planned trajectory comprises planned positions of the autonomous vehicle at a plurality of time points within a preset time period;the case that the conflict exists between the initially planned trajectories of the respective autonomous vehicles comprises: there being a time point at which the planned positions conflict in the plurality of time points within the preset time period.3. The vehicle trajectory planning method of claim 1 , wherein adjusting the planned trajectory of the autonomous vehicle associated with the conflict comprises:determining a modified planned trajectory for the autonomous vehicle associated with the conflict, and sending the modified planned trajectory to the autonomous vehicle associated with the conflict; and/or,sending an instruction for re-determining a planned trajectory to the autonomous vehicle associated with the conflict.4. The vehicle trajectory planning method of claim 2 , wherein adjusting the planned trajectory of the autonomous vehicle associated with the conflict comprises:determining a modified planned trajectory for the autonomous vehicle associated with the conflict, ...

Contingency Planning and Safety Assurance

A method for contingency planning for an autonomous vehicle (AV) includes determining a nominal trajectory for the AV; detecting a hazard object that does not intrude into a path of the AV at a time of the detecting the hazard object; determining a hazard zone for the hazard object; determining a time of arrival of the AV at the hazard zone; determining a contingency trajectory for the AV; controlling the AV according to the contingency trajectory; and, in response to the hazard object intruding into the path of the AV, controlling the AV to perform a maneuver to avoid the hazard object. The contingency trajectory includes at least one of a lateral contingency or a longitudinal contingency. The contingency trajectory is determined using the time of arrival of the AV at the hazard zone.

EFFICIENT INFERENCE UPDATE USING BELIEF SPACE PLANNING

An autonomous system comprising: at least one hardware processor; a sensors module; and a non-transitory computer-readable storage medium having stored thereon program instructions, the program instructions executable by the at least one hardware processor to: receive, from said sensors module, a set of measurements associated with a joint state of said autonomous system, infer, based, at least in part, on said set of measurements, a current belief regarding said joint state of said autonomous system, determine a control action based on said inference, wherein said determining comprises calculating a future belief regarding a future joint state of said autonomous system, wherein said future joint state is as a result of said control action, execute said control action, and generate a new inference based, at least in part, on said future belief, wherein said future belief is updated based on a new set of measurements from said sensors module. 1. An autonomous system comprising:at least one hardware processor;a sensors module; and receive, from said sensors module, a set of measurements associated with a joint state of said autonomous system,', 'infer, based, at least in part, on said set of measurements, a current belief regarding said joint state of said autonomous system,', 'determine a control action based on said inference, wherein said determining comprises calculating a future belief regarding a future joint state of said autonomous system, wherein said future joint state is as a result of said control action,', 'execute said control action, and', 'generate a new inference based, at least in part, on said future belief, wherein said future belief is updated based on a new set of measurements from said sensors module., 'a non-transitory computer-readable storage medium having stored thereon program instructions, the program instructions executable by the at least one hardware processor to2. The autonomous system of claim 1 , wherein said control action ...

Full Uncertainty for Motion Planning in Autonomous Vehicles

Systems and methods for motion planning by a vehicle computing system of an autonomous vehicle are provided. The vehicle computing system can input sensor data to a machine-learned system including one or more machine-learned models. The computing system can obtain, as an output of the machine-learned model(s), motion prediction(s) associated with object(s) detected by the system. The system can convert a shape of the object(s) into a probability of occupancy by convolving an occupied area of the object(s) with a continuous uncertainty associated with the object(s). The system can determine a probability of future occupancy of a plurality of locations in the environment at future times based at least in part on the motion prediction(s) and the probability of occupancy of the object(s). The system can provide the motion prediction(s) and the probability of future occupancy of the plurality of locations to a motion planning system of the autonomous vehicle. 1. A computer-implemented method of motion planning for autonomous driving , comprising:inputting, by a computing system comprising one or more computing devices, sensor data to a machine-learned system comprising one or more machine-learned models configured for object prediction in association with an environment external to an autonomous vehicle;obtaining, by the computing system as an output of the one or more machine-learned models, data indicative of one or more motion predictions associated with one or more objects detected by the machine-learned system;converting, by the computing system, a shape of the one or more objects detected by the machine-learned system into a probability of occupancy by convolving an occupied area of the one or more objects with a continuous uncertainty associated with the one or more objects;determining, by the computing system, data indicative of a probability of future occupancy of a plurality of locations in the environment at one or more future times based at least in part on ...

Lane change detection system and method for an autonomous vehicle

An autonomous vehicle configured to perform a lane change maneuver is described herein. The autonomous vehicle includes several different types of sensor systems, such as image, lidar, radar, sonar, infrared, and GPS. The autonomous vehicle additionally includes a computing system that executes instructions on a lane change detection system and a control system. The lane change detection system includes a region of interest module and an object trajectory module for determining whether the autonomous vehicle will collide with another object if the autonomous vehicle maneuvers into an adjacent lane. An instruction module of the lane change detection system is further used to facilitate operational control of a mechanical system of the autonomous vehicle, such as an engine or a steering system.

SYSTEM AND METHOD FOR IMPLEMENTING REWARD BASED STRATEGIES FOR PROMOTING EXPLORATION

A system and method for implementing reward based strategies for promoting exploration that include receiving data associated with an agent environment of an ego agent and a target agent and receiving data associated with a dynamic operation of the ego agent and the target agent within the agent environment. The system and method also include implementing a reward function that is associated with exploration of at least one agent state within the agent environment. The system and method further include training a neural network with a novel unexplored agent state. 1. A computer-implemented method for implementing reward based strategies for promoting exploration comprising:receiving data associated with an agent environment of an ego agent and a target agent;receiving data associated with a dynamic operation of the ego agent and the target agent within the agent environment;implementing a reward function that is associated with exploration of at least one agent state within the agent environment, wherein the reward function includes assigning at least one reward based on if the at least one agent state is a novel unexplored agent state or a previously explored agent state; andtraining a neural network with the novel unexplored agent state, wherein at least one simulation is processed to determine at least one additional novel unexplored agent state based on an analysis of at least one reward of the reward function.2. The computer-implemented method of claim 1 , wherein receiving data associated with the agent environment includes receiving image data and LiDAR data from at least one of the: ego agent and the target agent claim 1 , wherein the image data and the LiDAR data are fused to determine a simulated agent environment model that pertains to the agent environment at a current time step.3. The computer-implemented method of claim 2 , wherein receiving data associated with the agent environment includes receiving dynamic data associated with the ego agent and ...

Systems and methods for navigating a vehicle

Systems and methods are provided for vehicle navigation. In one implementation, a system may comprise an interface to obtain sensing data of an environment of the host vehicle. The processing device may be configured to determine a planned navigational action; identify, a target vehicle in the environment of the host vehicle; predict a distance between the host vehicle and the target vehicle if the planned navigational action was taken; determine a current host vehicle stopping distance based on a braking capability, acceleration capability, and speed of the host vehicle; determine a current target vehicle braking distance based on a speed and braking capability of the target vehicle; and implement the planned navigational action when the predicted distance of the planned navigational action is greater than a minimum safe longitudinal distance calculated based on the current host vehicle stopping distance and the current target vehicle braking distance.

DETECTING OUT-OF-MODEL SCENARIOS FOR AN AUTONOMOUS VEHICLE

Detecting out-of-model scenarios for an autonomous vehicle including: determining, based on first sensor data from one or more sensors, an environmental state relative to the autonomous vehicle, wherein operational commands for the autonomous vehicle are based on a selected machine learning model, wherein the selected machine learning model comprises a first machine learning model; comparing the environmental state to a predicted environmental state relative to the autonomous vehicle; and determining, based on a differential between the environmental state and the predicted environmental state, whether to select a second machine learning model as the selected machine learning model. 1. A method for detecting out-of-model scenarios for an autonomous vehicle , comprising:determining, based on first sensor data from one or more sensors, an environmental state relative to the autonomous vehicle, wherein operational commands for the autonomous vehicle are based on a selected machine learning model, wherein the selected machine learning model comprises a first machine learning model;comparing the environmental state to a predicted environmental state relative to the autonomous vehicle; anddetermining, based on a differential between the environmental state and the predicted environmental state, whether to select a second machine learning model as the selected machine learning model.2. The method of claim 1 , further comprising determining the predicted environmental state based on second sensor data.3. The method of claim 2 , wherein the second sensor data comprises sensor data associated with a time window ending at a time offset relative to a current time.4. The method of claim 1 , wherein determining claim 1 , based on the differential between the environmental state and the predicted environmental state claim 1 , whether to select the second machine learning model as the selected machine learning model comprises determining to select the second machine learning model ...

CONDITIONAL AGENT TRAJECTORY PREDICTION

Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for performing a conditional behavior prediction for one or more agents. The system obtains context data characterizing an environment. The context data includes data characterizing a plurality of agents, including a query agent and one or more target agents, in the environment at a current time point. The system further obtains data identifying a planned future trajectory for the query agent after the current time point, and for each target agent in the set, processes the context data and the data identifying the planned future trajectory using a first neural network to generate a conditional trajectory prediction output that defines a conditional probability distribution over possible future trajectories of the target agent after the current time point given that the query agent follows the planned future trajectory for the query agent after the current time point.

METHOD AND APPARATUS FOR AUTONOMOUS DRIVING CONTROL, ELECTRONIC DEVICE, AND STORAGE MEDIUM

The present application discloses a method and an apparatus for autonomous driving control, an electronic device, and a storage medium; the application relates to the technical field of autonomous driving. A specific implementation solution is: obtaining movement data of a pedestrian, where the movement data includes a velocity component of the pedestrian along a width direction of a lane and a time of duration that the pedestrian cuts into a driving path of the autonomous vehicle from one side; determining a movement direction of the pedestrian according to the movement data and the movement information of the pedestrian; and generating a driving strategy for the autonomous vehicle according to the movement direction of the pedestrian. Therefore, the movement direction of the pedestrian can be accurately predicted, which facilitates the autonomous vehicle to avoid the pedestrian and insures driving safety. 1. A method for autonomous driving control , comprising:obtaining movement data of a pedestrian; wherein the movement data comprises: a velocity component of the pedestrian along a width direction of a lane and a time of duration that the pedestrian cuts into a driving path of an autonomous vehicle from one side;determining a movement direction of the pedestrian according to the movement data and movement information of the pedestrian; andgenerating a driving strategy for the autonomous vehicle according to the movement direction of the pedestrian.2. The method according to claim 1 , wherein the obtaining movement data of a pedestrian comprises:obtaining a position of the lane where the pedestrian is located and a movement velocity of the pedestrian;resolving the movement velocity of the pedestrian in accordance to an extension direction and a width direction of the lane to obtain a velocity component along the extension direction of the lane and a velocity component along the width direction of the lane; anddetermining the time of duration that the pedestrian ...

VEHICLE-MOUNTED TERMINAL-BASED TRAFFIC ACCIDENT JUDGEMENT METHOD AND SYSTEM

The present disclosure provides a vehicle-mounted terminal-based traffic accident judgement method and system. The vehicle-mounted terminal-based traffic accident judgement method includes: transmitting an early-warning signal; judging whether a cancellation warning signal is received within a first preset time period; when no cancellation warning signal is received within the first preset time period, obtaining third-party information; judging whether in safe state according to the third-party information; transmitting a warning signal to a traffic accident monitoring server when in unsafe state. 1. A vehicle-mounted terminal-based traffic accident judgement method , comprising:transmitting an early-warning signal;judging whether a cancellation warning signal is received within a first preset time period;when no cancellation warning signal is received within the first preset time period, obtaining third-party information;judging whether in safe state according to the third-party information;transmitting a warning signal to a traffic accident monitoring server when in unsafe state.2. The method according to claim 1 , wherein the obtaining third-party information and judging whether in safe state according to the third-party information claim 1 , includes:receiving vehicle malfunction information of a vehicle transmitted by a vehicle-mounted automatic diagnosis device; andjudging whether the vehicle is in safe state according to the vehicle malfunction information.3. The method according to claim 2 , wherein after judging whether the vehicle is in safe state according to the vehicle malfunction information claim 2 , the obtaining third-party information and judging whether in safe state according to the third-party information claim 2 , further includes:when the vehicle is in safe state, obtaining face information of a driver;judging whether the driver is in safe state according to the face information of the driver.4. The method according to claim 1 , wherein before ...

COLLISION AVOIDANCE PERCEPTION SYSTEM

A collision avoidance system may validate, reject, or replace a trajectory generated to control a vehicle. The collision avoidance system may comprise a secondary perception component that may comprise one or more machine learned models, each of which may be trained to output one or more occupancy maps based at least in part on sensor data of different types. The occupancy maps may include a prediction of whether at least a portion of an environment is occupied at a future time by any one of multiple object types. Occupancy maps associated with a same time may be aggregated into a data structure that may be used to validate, reject, or replace the trajectory. 1two or more sensors;one or more processors; and receiving first sensor data associated with a first sensor type associated with at least one of the two or more sensors;', 'receiving second sensor data associated with a second sensor type associated with at least one of the two or more sensors;', 'determining, based at least in part on the first sensor data, a first current occupancy map and a first predicted occupancy map, the first predicted occupancy map comprising a discretized grid indicative of likelihoods that respective portions thereof are occupied;', the first current occupancy map and the second current occupancy map indicate a first likelihood and a second likelihood that a portion of an environment is occupied at a current time; and', 'the first predicted occupancy map and the second predicted occupancy map indicate a third likelihood and a fourth likelihood that the portion of the environment is occupied at a future time;, 'determining, based at least in part on the second sensor data, a second current occupancy map and a second predicted occupancy map, wherein, 'combining the first current occupancy map and the second current occupancy map into a data structure indicating whether the portion of the environment is occupied or unoccupied at the current time;', 'combining the first predicted ...

NAVIGATING MULTI-WAY STOP INTERSECTIONS WITH AN AUTONOMOUS VEHICLE

The subject matter described in this specification is directed to a system and techniques for operating an autonomous vehicle (AV) at a multi-way stop intersection. After detecting the AV is at a primary stopline of the multi-way stop intersection, a planned travel path though the multi-way stop intersection is obtained. If the planned travel path of the AV through the multi-way stop intersection satisfies a set of one or more clearance criteria, the AV proceeds past the primary stopline. The clearance criteria include a criterion that is satisfied in response to detecting the AV is clear to safely merge into a travel lane corresponding to the planned travel path. 1. A system comprisingone or more computer processors; andone or more non-transitory storage media storing instructions which, when executed by the one or more computer processors, cause performance of operations comprising: detecting, using a processing circuit, movement of a second vehicle at the intersection, the second vehicle having an expected travel path through the intersection that intersects a planned travel path of the first vehicle through the intersection; and', 'in accordance with a determination, based on the detected movement of the second vehicle, that the second vehicle is expected to exit the intersection, instructing, using a control circuit, the first vehicle to proceed into the intersection before the second vehicle exits the intersection., 'while a first vehicle is operating in an autonomous mode at a multi-way stop intersection and has a highest precedence at the multi-way stop intersection2. The system of claim 1 , wherein the determination claim 1 , based on the detected movement of the second vehicle claim 1 , that the second vehicle is expected to exit the intersection includes determining the trajectory of the second vehicle.3. The system of claim 2 , wherein the instructions further cause performance of operations comprising:in accordance with a determination that the ...

Onboard cluster tracking system

The technology relates to tracking objects in an environment around an autonomous vehicle. A computing system of the autonomous vehicle determines accurate motion characteristics of objects detected in its environment despite various sensor measurement limitations. By correcting motion distortion for fast moving objects and accounting for discrepancies in sensor data gathering, motion characteristics may be determined for the detected objects with enhanced accuracy. Multiple sets of correspondences are determined for clusters from multiple sensor spins, enabling better alignment using a surface matching algorithm even when clusters have fewer data points. Efficiency is also enhanced by selecting hypotheses based on confidence levels. These techniques provide for identifying the types of objects for which a yaw rate can be accurately determined. Object classification can also be improved by accumulating associated clusters corresponding to a detected object. In addition, under- or over-segmentation can be mitigated with such techniques.

TARGET ARRANGEMENT, METHOD, AND CONTROL UNIT FOR FOLLOWING A TARGET VEHICLE

Method, control unit, and target arrangement of a leading vehicle for triggering a follower vehicle, which is situated at a lateral distance from the leading vehicle, to coordinate its movements with the leading vehicle. The target arrangement comprises a target configured to be placed at a lateral distance from to the leading vehicle. The target is also configured to be recognized by at least one forwardly directed sensor of the follower vehicle. 1. A target arrangement of a leading vehicle for triggering a follower vehicle , situated at a lateral distance from the leading vehicle , to coordinate its movements with the leading vehicle , wherein the target arrangement comprises: a target configured to be placed at a lateral distance from the leading vehicle and configured to be recognized by at least one forwardly directed sensor of the follower vehicle.2. The target arrangement according to claim 1 , wherein the target comprises information concerning:a lateral distance between the target and the sensor of the follower vehicle for the follower vehicle to keep;a longitudinal distance to keep between the target and the sensor of the follower vehicle;vehicle path alignment information for the follower vehicle to keep; and/ora speed for the follower vehicle to keep.3. The target arrangement according to claim 1 , wherein the target comprises a plurality of target portions claim 1 , which are to be aligned with the sensor in the follower vehicle.4. The target arrangement according to claim 1 , wherein the target comprises a plurality of target portions claim 1 , each associated with a respective longitudinal distance to keep between the target and the sensor of the follower vehicle claim 1 , wherein the target portion to be presented is selectable from the leading vehicle.5. The target arrangement according to claim 1 , further comprising: a lateral distance between the target and the sensor of the follower vehicle to keep by the follower vehicle;', 'a longitudinal ...

DETERMINING PIXELS BEYOND NOMINAL MAXIMUM SENSOR DEPTH

Sensors, including time-of-flight sensors, may be used to detect objects in an environment. In an example, a vehicle may include a time-of-flight sensor that images objects around the vehicle, e.g., so the vehicle can navigate relative to the objects. The sensor may generate first image data at a first configuration and second image data at a second configuration. A disambiguated depth of a surface may be determined from the first image data and the second image data. If the disambiguated depth is greater than a nominal maximum depth of the sensor in the first configuration, an intensity of the surface may be determined from the first image data. If the intensity meets or exceeds a threshold intensity, the surface may be determined to be beyond the nominal maximum depth. If the intensity is less than the threshold intensity, an actual depth of the surface may be determined form the second image data as a distance less than the nominal maximum depth. 1. A vehicle comprising:a time-of-flight sensor;one or more processors; and receiving first depth information and first intensity information generated by the time-of-flight sensor, the first depth information being based, at least in part, on a first modulation frequency of the time-of-flight sensor;', 'receiving second depth information and second intensity information generated by the time-of-flight sensor, the second depth information being based, at least in part, on a second modulation frequency higher than the first modulation frequency;', 'determining, based at least in part on the first depth information, a plurality of first candidate depths for a surface;', 'determining, based at least in part on the second depth information, a plurality of second candidate depths for the surface;', 'determining, based on the plurality of first candidate depths and the plurality of second candidate depths, a disambiguated depth for the surface;', 'determining that the disambiguated depth is greater than a nominal maximum ...

DEPTH DEPENDENT PIXEL FILTERING

Sensors, including time-of-flight sensors, may be used to detect objects in an environment. In an example, a vehicle may include a time-of-flight sensor that images objects around the vehicle, e.g., so the vehicle can navigate relative to the objects. Sensor data generated by the time-of-flight sensor can include returns associated with highly reflective objects that cause glare. In some examples, a depth of a sensed surface is determined from the sensor data and additional pixels at the same depth are identified. The subset of pixels at the depth are filtered by comparing a measured intensity value to a threshold intensity value for the depth. Other threshold intensity values can be applied to subsets of pixels at different depths. 1. A vehicle comprising:a time-of-flight sensor configured to generate data based on light received at a receiver of the time-of-flight sensor;one or more processors; and receiving first sensor data from the time-of-flight sensor, the first sensor data comprising first depth information and first intensity information for a plurality of pixels generated with the time-of-flight sensor in a first configuration having a first modulation frequency;', 'receiving second sensor data from the time-of-flight sensor, the second sensor data comprising second depth information and second intensity information for the plurality of pixels generated with the time-of-flight sensor in a second configuration having a second modulation frequency higher than the first modulation frequency;', 'determining, based at least in part on the first depth information and the second depth information, a pixel of the plurality of pixels having a first depth that is greater than a nominal maximum depth of the time-of-flight sensor in the first configuration;', 'determining, based on the first sensor data and the second sensor data, additional pixels of the plurality of pixels having the first depth, the pixel and the additional pixels comprising a first subset of the ...

REMOTE SUPPORT SYSTEM AND REMOTE SUPPORT METHOD

A remote support system is configured to determine whether a vehicle collides with an object to be avoided when the vehicle gets into a remote control request situation, stop to send a remote control request when the remote support system determines that the vehicle does not collide with the object, generate a first speed plan for the vehicle to continue autonomous driving at a predicted collision position and a second speed plan for the vehicle to stop before reaching the predicted collision position when the remote support system determines that the vehicle collides with the object, and determine to send a remote control request based on the degree of deviation between these speed plans. 1. A remote support system configured to send a remote control request to a remote operator when a vehicle traveling autonomously gets into a remote control request situation , the remote support system comprising:a storage device storing at least one program; andat least one processor connected to the storage device, wherein:the at least one processor is configured to perform a first determination process and a second determination process by executing the at least one program when the vehicle gets into the remote control request situation,the first determination process is a process of determining whether the vehicle collides with an object to be avoided regarding the remote control request situation,the second determination process is a process of determining necessity of sending the remote control request based on a result of the first determination process, acquiring at least one of map information around the vehicle, surrounding environment information regarding a surrounding environment of the vehicle, and vehicle motion information regarding motion of the vehicle, and', 'determining whether the vehicle collides with the object to be avoided, based on at least one of the map information, the vehicle motion information, and the surrounding environment information; and, 'the ...

ASSESSING PRESENT INTENTIONS OF AN ACTOR PERCEIVED BY AN AUTONOMOUS VEHICLE

Methods of forecasting intentions of actors that an autonomous vehicle (AV) encounters in are disclosed. The AV uses the intentions to improve its ability to predict trajectories for the actors, and accordingly making decisions about its own trajectories to avoid conflict with the actors. To do this, for any given actor the AV determines a class of the actor and detects an action that the actor is taking. The system uses the class and action to identify candidate intentions of the actor and evaluating a likelihood of each candidate intention. The system repeats this process over multiple cycles to determine overall probabilities for each of the candidate intentions. The AV's motion planning system can use the probabilities to determine likely trajectories of the actor, and accordingly influence the trajectory that the AV will itself follow in the environment. 1. A method of forecasting an intention of an actor in an environment through which an autonomous vehicle is traveling , the method comprising: detecting an actor that is proximate to the autonomous vehicle,', 'determining a class of the actor, and', 'detecting an action that the actor is taking;, 'by a perception system of an autonomous vehicle using the class and the detected action to generate a plurality of candidate intentions of the actor,', 'evaluating a likelihood of each candidate intention, and', 'saving each of the candidate intentions and their likelihoods in a data store;, 'by a forecasting system of the autonomous vehicle for each cycle of a plurality of cyclesafter any current cycle of the plurality of cycles has completed, analyzing the candidate intentions and their likelihoods for the current cycle and for one or more of the prior cycles to determine an overall probability for each of the candidate intentions; andby a motion planning system of the autonomous vehicle, using the overall probabilities to select one of the candidate intentions to influence a selected trajectory for the autonomous ...

Two-step reference line smoothing method to mimic human driving behaviors for autonomous driving cars

A method of determining a smooth reference line for navigating an autonomous vehicle in a manner similar to human driving is disclosed. A high density map is used to generate a centerline for a lane of roadway. Using the centerline, a number of sample points is generated that is related to a curvature of the centerline. Adjustment points are generated at each sample point, a few on either side of the centerline at each sample point. Candidate points at a sample point include the adjustment points and sample point. A least cost path is determined through each of the candidate points at each of the sample points. Path cost is based an angle of approach and departure through a candidate point, and a distance of the candidate point from the centerline.

Autonomous Vehicle Motion Control Systems and Methods

Systems and methods for controlling the motion of an autonomous are provided. In one example embodiment, a computer-implemented method includes obtaining data associated with an object within a surrounding environment of an autonomous vehicle. The data associated with the object is indicative of a predicted motion trajectory of the object. The method includes determining a vehicle action sequence based at least in part on the predicted motion trajectory of the object. The vehicle action sequence is indicative of a plurality of vehicle actions for the autonomous vehicle at a plurality of respective time steps associated with the predicted motion trajectory. The method includes determining a motion plan for the autonomous vehicle based at least in part on the vehicle action sequence. The method includes causing the autonomous vehicle to initiate motion control in accordance with at least a portion of the motion plan.

PROCESS TO LEARN NEW IMAGE CLASSES WITHOUT LABELS

Described is a system for learning object labels for control of an autonomous platform. Pseudo-task optimization is performed to identify an optimal pseudo-task for each source model of one or more source models. An initial target network is trained using the optimal pseudo-task. Source image components are extracted from source models, and an attribute dictionary of attributes is generated from the source image components. Using zero-shot attribution distillation, the unlabeled target data is aligned with the source models similar to the unlabeled target data. The unlabeled target data are mapped onto attributes in the attribute dictionary. A new target network is generated from the mapping, and the new target network is used to assign an object label to an object in the unlabeled target data. The autonomous platform is controlled based on the object label. 1. A system for learning object labels for control of an autonomous platform , the system comprising: performing pseudo-task optimization to identify an optimal pseudo-task for each source model of one or more source models;', 'training an initial target network with self-supervised learning using the optimal pseudo-task;', 'extracting a plurality of source image components from the one or more source models;', 'generating an attribute dictionary of abstract attributes from the plurality of source image components;', 'using zero-shot attribution distillation, aligning a set of unlabeled target data with the one or more source models that are similar to the set of unlabeled target data;', 'mapping the set of unlabeled target data onto a plurality of abstract attributes in the attribute dictionary;', 'generating a new target network from the mapping;', 'using the new target network, assigning an object label to an object in the unlabeled target data; and', 'controlling the autonomous platform based on the assigned object label., 'one or more processors and a non-transitory computer-readable medium having ...

Multi-point enforced based stitch method to connect two smoothed reference lines

In one embodiment, a method for generating a reference line for operating an autonomous driving vehicle includes determining a first ending reference point having a smallest curvature among a plurality of points within a first defined distance along a path, generating a first reference line based on a first initial reference point and the first ending reference point, determining a second ending reference point having a smallest curvature among a plurality of points within a second defined distance along the path, generating a second reference line based on the first and second ending reference points and an end section of the first reference line, connecting the first and second reference lines, and controlling the autonomous driving vehicle along the connected first reference line and the second reference line.

Control of a Motor Vehicle

A method comprises the steps of detecting a first trajectory of a first motor vehicle while traveling a predetermined route; determining a course of the route based on predetermined map data; determining a quality to which the trajectory and the course correspond; and transmitting the quality to a second motor vehicle. 1. A method comprising the following steps:detecting a first trajectory of a first motor vehicle while traveling a predetermined route;determining a course of the route on the basis of predetermined map data;determine a quality to which the first trajectory and the course correspond;transmitting the quality to a second motor vehicle.2. The method according to claim 1 , wherein a plurality of first trajectories of first motor vehicles are detected and the quality is determined with respect to the plurality of first trajectories.3. The method according to claim 2 , wherein the quality is determined with respect to first trajectories which were detected when traveling the route within a predetermined past time period.4. The method according to claim 1 , wherein a distance is determined between the first trajectory and the course and the quality is the higher the smaller the distance.5. The method according to claim 1 , wherein the first trajectory comprises a number of positions of the first motor vehicle and the higher a density of positions of first motor vehicles claim 1 , the higher the quality.6. The method according to claim 5 , wherein the quality is the lower claim 5 , the stronger the curvature of the route at the same density of positions determining the route.7. A method comprising the following steps:determining a course of a route on the basis of predetermined map data;detecting a quality assigned to the course;scanning of an environment of a second motor vehicle;determining a second trajectory by merging the course with the scan; andcontrolling the second motor vehicle to follow the second trajectory.8. The method according to claim 7 , ...

Method, Computer Program Product, and Driver Assistance System for Determining One or More Lanes of a Road in an Environment of a Vehicle

A method determines one or more lanes of a road in an environment of a vehicle, by receiving a plurality of objects in the environment of the vehicle; receiving a plurality of trajectories of the plurality of objects in the environment of the vehicle; estimating a shape of a road based on the plurality of trajectories of the plurality of objects; and determining one or more lanes of the road using the estimated shape of the road and the plurality of objects and/or the plurality of trajectories of the plurality of objects. 1. A method for determining one or more lanes of a road in an environment of a vehicle , the method comprising:receiving a plurality of objects in the environment of the vehicle;receiving a plurality of trajectories for the plurality of objects in the environment of the vehicle;estimating a shape of the road based on the plurality of trajectories for the plurality of objects; anddetermining the one or more lanes of the road using the estimated shape of the road and the plurality of objects and/or the plurality of trajectories of the plurality of objects.2. The method according to claim 1 , whereinthe plurality of trajectories comprises a single trajectory for each object of at least a subset of objects of the plurality of objects.3. The method according to claim 1 , wherein the step of estimating the shape of the road based on the plurality of trajectories comprises:determining a number of segments for the plurality of trajectories;clustering the plurality of the trajectories in each of the one or more segments according to one or more shapes of trajectories of the plurality of trajectories;determining a cluster of trajectories of the clustered plurality of trajectories in each of the one or more segments, wherein the cluster of trajectories comprises a majority of trajectories having an equal or similar shape in a particular segment of the one or more segments; andestimating the shape of the road based on the determined cluster of trajectories in ...

Nearby Driver Intent Determining Autonomous Driving System

An autonomous driving system capable of determining an intent of a nearby human driver and taking an action to avoid a collision is presented. The system may receive a current state of a nearby vehicle, determine an expected action of a human driver of the nearby vehicle by determining a result of a reward function, the reward function being a linear combination of feature functions, where each feature function is a neural network which has been trained to reproduce a corresponding algorithmic feature function, and based on the determined expected action of the human driver, taking an action to avoid a collision. 1. A method comprising:receiving, by a computing device in a first vehicle, a current state of a second vehicle;based on the current state, determining an expected action of a human driver of the second vehicle by determining a result of a reward function, wherein the reward function comprises a linear combination of feature functions, the feature functions having corresponding weights, wherein each feature function comprises a neural network which has been trained to reproduce a corresponding algorithmic feature function; andbased on the determined expected action of the human driver, communicating with a vehicle control interface of the first vehicle to cause the first vehicle to take a mitigating action to avoid a collision.2. The method of claim 1 , wherein the receiving the current state of the second vehicle comprises receiving the current state of the second vehicle from a camera in the first vehicle.3. The method of claim 1 , wherein the algorithmic feature function comprises a function for keeping a speed claim 1 , collision avoidance claim 1 , keeping a heading claim 1 , or maintaining a lane boundary distance.4. The method of claim 1 , wherein the weights are resultant from preference-based learning of the reward function with human subjects.5. The method of claim 4 , wherein each neural network has been further trained on results from the ...

Method for Creating a Probabilistic Free Space Map with Static and Dynamic Objects

The invention relates to a method for creating a probabilistic free space map with static () and dynamic objects (V-V), having the following steps: 122317ab. A method for creating a probabilistic free space map with static ( , , ) and dynamic objects (V-V) , having the following steps:{'b': 1', '2', '2', '3, 'i': a', 'b, 'retrieving (S) static objects (, , ) as well as a perception area polygon (WP) from an existing environment model;'}{'b': 2', '1', '2', '1', '7, 'collecting (S) predicted trajectories (T, T) of dynamic objects (V-V);'}{'b': 3', '2', '2', '3', '1', '2, 'i': a', 'b, 'merging (S) the static objects (, , ) of the perception area polygon (WP) and the predicted trajectories (T, T) in a first free space map;'}{'b': '4', 'fixing (S) a maximum prediction time;'}{'b': '5', 'fixing (S) prediction time steps;'}{'b': '6', 'fixing (S) a current prediction time and setting this current prediction time to the value 0 in order to fix the start of a fixed prediction time period;'}{'b': 7', '2', '2', '3', '1', '7, 'i': a', 'b, 'fixing (S) confidence regions (K) around the static (, , ) and dynamic objects (V-V);'}{'b': 8', '2', '2', '3', '1', '7, 'i': a', 'b, 'fixing (S) at least one uncertain region (U) around at least one static (, , ) or dynamic object (V-V);'}{'b': '9', 'producing (S) a first probabilistic free space map for the current prediction time;'}{'b': '10', 'producing (S) at least one further free space map for at least one prediction time step;'}{'b': '11', 'evaluating (S) the produced free space maps.'}2. The method according to claim 1 , characterized in that a free space map is produced for each prediction time step until the maximum prediction time is reached.317. The method according to claim 1 , characterized in that the at least one uncertain region (U) is fixed based on the existing environment model and the trajectory prediction of the dynamic objects (V-V).41217. The method according to claim 1 , characterized in that the at least one ...

Collision avoidance system with trajectory validation

A vehicle may include a primary system and a secondary system to validate operation of the primary system and to control the vehicle to avoid collisions. For example, the secondary system may receive multiple trajectories from the primary system, such as a primary trajectory and a secondary, contingent, trajectory associated with a deceleration or other maneuver. The secondary system may determine if a trajectory is associated with a potential collision, if the trajectory is consistent with a current or previous pose, if the trajectory is compatible with a capability of the vehicle, etc. The secondary system may select the primary trajectory if valid, the secondary trajectory if the primary trajectory is invalid, or another trajectory generated by the secondary system if the primary trajectory and the secondary trajectory are invalid. If no valid trajectory is determined, the vehicle may decelerate at a maximum rate.

METHOD AND APPARATUS FOR PREVENTING ESCAPE OF AUTONOMOUS VEHICLE

A moving object escape prevention method includes: controlling, by a processor of a moving object, to drive the moving object based on autonomous driving; detecting, by the processor, whether a collision occurred by the moving object; in response to detecting the collision, transmitting, by the processor, a collision occurrence notification signal and position information of the moving object to an Intelligent Transportation System Infrastructure (ITSI); receiving, by the processor, escape-related information from the ITSI. The receiving escape-related information includes: determining, by the ITSI, whether or not the moving object escapes based on position information of the moving object; receiving, by the processor, accident handling information from the ITSI upon determining that the moving object does not escape, and receiving, by the processor, an escape warning message from the ITSI when the position information of the moving object changes. 1. A moving object escape prevention method comprising:controlling, by a processor of a moving object, to autonomously drive the moving object;detecting, by the processor, whether a collision occurred by the moving object;in response to detecting the collision, transmitting, by the processor, a collision occurrence notification signal and position information of the moving object to an Intelligent Transportation System Infrastructure (ITSI);receiving, by the processor, escape-related information from the ITSI, determining, by the ITSI, whether or not the moving object escapes based on position information of the moving object;', 'receiving, by the processor, accident handling information from the ITSI upon determining that the moving object does not escape, and', 'receiving, by the processor, an escape warning message from the ITSI when the position information of the moving object changes., 'wherein the receiving escape-related information includes2. The moving object escape prevention method of claim 1 , wherein the ...

SYSTEM FOR GRID-BASED MERGE CUT-IN PREDICTION AND METHOD THEREOF

A prediction system implemented in a host vehicle to predict a merge cut-in for an autonomous vehicle. The system comprises an input unit for capturing neighboring information of the host vehicle, and a processing unit to receive the captured neighboring information and generate a grid map by determining shape and dimensions of a grid, estimate trajectory of each target vehicle of the one or more target vehicles, based on a driver behavior model of each target vehicle, to determine optimized path of each target vehicle, and generate a global maneuver model by analyzing motion of each neighboring target vehicle, wherein on generation of the global maneuver model a merge cut-in threat for the host vehicle is computed by performing centralized risk management and utilizing the predicted trajectory of the one or more target vehicles. 1. A prediction system implemented in a host vehicle , said system comprising:an input unit comprising one or more sensors to capture neighboring information of the host vehicle, the neighboring information comprising information of one or more target vehicles and information of surroundings, in proximity of the host vehicle, wherein the host vehicle is positioned in a lane of a road; and receive the captured neighboring information from the input unit;', 'generate a grid map by determining dimensions of a grid based on analysis of any or a combination of the captured neighboring information and one or more attributes of the host vehicle;', 'estimate trajectory of each target vehicle of the one or more target vehicles, based on a driver behavior model of each target vehicle, to determine optimized path of each target vehicle, wherein dimensions of the grid are updated based on said estimation; and', 'generate a global maneuver model by analyzing motion of each target vehicle based selection of a model classifier for each target vehicle, wherein the model classifier is selected by analyzing the predicted trajectory of each target vehicle, ...

All Mover Priors

Systems, devices, products, apparatuses, and/or methods for generating a driving path for an autonomous vehicle on a roadway by determining one or more prior probability distributions of one or more motion paths for one or more objects that have previously moved in a geographic location and/or for controlling travel of an autonomous vehicle on a roadway by predicting movement of a detected object according to one or more prior probability distributions of one or more motion paths for one or more objects that have previously moved in a geographic location. 1. A computer-implemented method comprising:obtaining, with a computing system comprising one or more processors, one or more prior probability distributions of one or more motion paths for one or more objects that previously moved in a geographic location;obtaining, with the computing system, sensor data associated with a detected object in an environment surrounding an autonomous vehicle;determining, with the computing system, one or more prediction scores based on the one or more prior probability distributions and the sensor data, wherein the one or more prediction scores include one or more predictions of whether the detected object is moving over at least one motion path of the one or more motion paths; andcontrolling, with the computer system, travel of the autonomous vehicle on a roadway based on the one or more prediction scores.2. The computer-implemented method of claim 1 , wherein the one or more prior probability distributions are associated with one or more probability values that correspond to one or more elements of a plurality of elements in a map of the geographic location claim 1 , wherein the one or more probability values include one or more probabilities of the one or more objects at one or more positions in the geographic location associated with the one or more elements in the map moving over the one or more motion paths.3. The computer-implemented method of claim 2 , wherein the one or more ...

METHOD FOR PLANNING PATH FOR LANE CHANGING, ELECTRONIC DEVICE AND COMPUTER READABLE MEDIUM

A method for planning a path for lane changing includes: determining, based on position information of a detected obstacle, a reference position to be passed by a vehicle when the vehicle detours the obstacle; estimating an end position of the vehicle detouring the obstacle based on a positional relationship between the vehicle and the obstacle; and determining a path of the vehicle to detour the obstacle based on a current position of the vehicle, the reference position and the end position. 1. A method for planning a path for lane changing , comprising:determining, based on position information of a detected obstacle, a reference position to be passed by a vehicle when the vehicle detours the obstacle;estimating an end position of the vehicle detouring the obstacle based on a positional relationship between the vehicle and the obstacle; anddetermining a path of the vehicle to detour the obstacle based on a current position of the vehicle, the reference position and the end position.2. The method of claim 1 , wherein claim 1 , before determining the reference position to be passed by the vehicle when the vehicle detours the detected obstacle based on the position information of the obstacle claim 1 , the method further comprises:detecting the position information of the obstacle and a motion state of the obstacle before the vehicle starts lane changing.3. The method of claim 1 , wherein determining the reference position to be passed by the vehicle when the vehicle detours the detected obstacle based on the position information of the obstacle comprises:obtaining a boundary position of the obstacle close to the vehicle from the position information of the obstacle; andcalculating a position with a predetermined deviation distance from the boundary position, to obtain the reference position to be passed by the vehicle when the vehicle detours the obstacle,wherein the predetermined deviation distance is a preset spacing distance between the obstacle and the vehicle ...

ELECTRONIC DEVICE FOR INTEGRATING AND PREDICTING FUTURE TRAJECTORIES OF UNSPECIFIED NUMBER OF SURROUNDING VEHICLES AND OPERATING METHOD THEREOF

An electronic device and an operating method thereof are for predicting future trajectories of an unspecified number of surrounding vehicles in an integrated way. The electronic device and operating method may be configured to recognize historical trajectories of one or more surrounding objects, predict future trajectories of the surrounding objects in an integrated way based on the recognized historical trajectories, and plan a driving trajectory of the electronic device based on the predicted future trajectories of the surrounding objects. 1. An operating method of an electronic device , comprising:recognizing historical trajectories of one or more surrounding objects;predicting future trajectories of the surrounding objects in an integrated way based on the recognized historical trajectories; andplanning a driving trajectory of the electronic device based on the predicted future trajectories of the surrounding objects.2. The operating method of claim 1 , wherein the predicting of the future trajectories comprises:configuring a graph model based on the recognized historical trajectories; andpredicting the future trajectories based on the graph model.3. The operating method of claim 2 ,wherein the graph model comprises a plurality of nodes and a plurality of edges connecting the nodes,wherein the nodes indicate the surrounding objects and the electronic device, respectively, andwherein the edges indicate interactions among the surrounding objects and between the electronic device and each of the surrounding objects, and directionality of the interactions, respectively.4. The operating method of claim 3 ,wherein edges connecting nodes indicating the surrounding objects contains bi-directionality, andwherein a edges connecting any one of the nodes indicating the surrounding object and a node indicating the electronic device contains uni-directionality directed from the electronic device to the surrounding object.5. The operating method of claim 3 , wherein the ...

Machine-learning based system for path and/or motion planning and method of training the same

A system and method for path and/or motion planning and for training such a system are described. In one aspect, the method comprises generating a sequence of predicted occupancy grid maps (OGMs) for T-T1 time steps based on a sequence of OGMs for 0-T1 time steps, a reference map of an environment in which an autonomous vehicle is operating, and a trajectory. A cost volume is generated for the sequence of predicted OGMs. The cost volume comprises a plurality of cost maps for T-T1 time steps. Each cost map corresponds to a predicted OGM in the sequence of predicted OGMs and has the same dimensions as the corresponding predicted OGM. Each cost map comprises a plurality of cells.Each cell in the cost map represents a cost of the cell in corresponding predicted OGM being occupied in accordance with a policy defined by a policy function.

PREDICTIVE TURNING ASSISTANT

A method for assisting in turning a vehicle, the method may include detecting or estimating that the vehicle is about to turn to a certain direction or is turning to the certain direction; sensing a relevant portion of an environment of the vehicle to provide sensed information, wherein the relevant portion of the environment is positioned at a side of the vehicle that corresponds with the certain direction; applying an artificial intelligence process on the sensed information to (i) detect objects within the relevant portion of the environment and (ii) estimate expected movement patterns of the objects within a time frame that ends with an expected completion of the turn of the vehicle; determining, given an expected trajectory of the vehicle during the turn and the expected movement patterns of the objects, whether at least one of the objects is expected to cross the trajectory of the vehicle during the turn; and responding to an outcome of the determining. 1. A method for assisting in turning a vehicle , the method comprises:detecting or estimating that the vehicle is about to turn to a certain direction or is turning to the certain direction;sensing a relevant portion of an environment of the vehicle to provide sensed information, wherein the relevant portion of the environment is positioned at a side of the vehicle that corresponds with the certain direction;applying an artificial intelligence process on the sensed information to (i) detect objects within the relevant portion of the environment and (ii) estimate expected movement patterns of the objects within a time frame that ends with an expected completion of the turn of the vehicle;determining, given an expected trajectory of the vehicle during the turn and the expected movement patterns of the objects, whether at least one of the objects is expected to cross the trajectory of the vehicle during the turn; andresponding to an outcome of the determining.2. The method according to claim 1 , wherein the ...

METHOD AND APPARATUS FOR ANNOTATING VIRTUAL LANE AT CROSSING

A method and apparatus for annotating a virtual lane at a crossing is provided, which relates to the field of intelligent transportation, and specifically includes: calculating a virtual connecting probability of various lanes that do not connected to each other at the crossing based on driving trajectory data of a vehicles that passing through the crossing within a detection time, and generating the virtual lane at the crossing for two disconnected lanes whose virtual connecting probability is greater than a threshold, and annotating the virtual lane on a map. In this process, the virtual lane at the crossing can be automatically generated according to the driving trajectory data of the vehicles passing through the crossing, and because the driving trajectory data of the vehicles passing through the crossing is real trajectory data, which is more in conformity with actual driving rules of the vehicles, and has a higher annotation accuracy. 1. A method for annotating a virtual lane at a crossing , wherein the method comprises:acquiring driving trajectory data of vehicles that passing through the crossing within a detection time; calculating a virtual connecting probability of any two disconnected lanes at the crossing according to the driving trajectory data of the vehicles that passing through the crossing; the any two disconnected lanes are any two lanes that have no real lane line connected to each other; the virtual connecting probability of the any two disconnected lanes is a probability that vehicles driving from one lane to the other between the any two disconnected lanes;aggregating driving trajectory data of two disconnected lanes whose virtual connecting probability is greater than a threshold when passing through the crossing, to generate the virtual lane at the crossing; andannotating the virtual lane on a map.2. The method according to claim 1 , wherein the calculating a virtual connecting probability of any two disconnected lanes at the crossing ...

VISUALIZATION OF PLANNED AUTONOMOUS VEHICLE BEHAVIOR

To visualize planned behavior of an autonomous vehicle (AV) traveling along a roadway, a user interface engine receives data describing a planned pathway of the AV along the roadway and object data describing an object having a predicted pathway crossing the planned pathway of the AV at a cross point. The user interface engine classifies the object either an asserting object or a yielding object based on a prediction of whether the object reaches the cross point before the AV or after the AV. The user interface engine generates an image that includes the planned pathway of the AV and the object in the environment of the AV. The image of the object indicates whether the object is classified as an asserting object or a yielding object. 1. A method for visualizing planned behavior of an autonomous vehicle (AV) , the method comprising:receiving data describing a planned pathway of the AV along a roadway;receiving object data describing an object in an environment of the AV, the object associated with a predicted pathway crossing the planned pathway of the AV at a cross point;classifying the object as one of an asserting object and a yielding object based on a prediction of whether the object reaches the cross point before the AV or after the AV; andgenerating an image comprising the planned pathway of the AV and the object in the environment of the AV, wherein the image of the object indicates whether the object is classified as an asserting object or a yielding object.2. The method of claim 1 , wherein claim 1 , in response to the object being an asserting object claim 1 , the image of the object has a first visual characteristic claim 1 , and in response to the object being a yielding object claim 1 , the image of the object has a second visual characteristic different from the first visual characteristic.3. The method of claim 2 , wherein the first visual characteristic is a first color claim 2 , and the second visual characteristic is a second color.4. The method of ...

COLLISION WARNING SYSTEM FOR SAFETY OPERATORS OF AUTONOMOUS VEHICLES

Embodiments disclose a system and method to send an alert/warning for a potential collision to a safety operator of an autonomous driving vehicle (ADV). According to one embodiment, a system perceives an environment of an autonomous driving vehicle (ADV), including one or more obstacles. The system determines whether the ADV will potentially collide with the one or more obstacles based on a planned trajectory. If the ADV is determined to potentially collide, the system determines a time to collision based on the planned trajectory and the one or more obstacles. If the determined time to collision is less than a threshold or the time to collision decreases for a predetermined number of consecutive planning cycles, the system generates a warning signal to alert an operator of the ADV. The system sends the warning signal to an operator interface of the ADV to alert the operator of the potential collision. 1. A computer-implemented method for operating an autonomous driving vehicle (ADV) , the method comprising:perceiving an environment for an autonomous driving vehicle (ADV), including one or more obstacles;determining whether the ADV will potentially collide with the one or more obstacles based on a planned trajectory;if the ADV is determined to potentially collide, determining a time to collision based on the planned trajectory and the one or more obstacles;if the determined time to collision is less than a threshold or the time to collision decreases for a predetermined number of consecutive planning cycles, generating a warning signal to alert an operator of the ADV; andsending the warning signal to a user interface of the ADV to alert the operator of the potential collision.2. The computer-implemented method of claim 1 , wherein the warning signal is sent through a controlled area network (CAN) bus to a user interface of the ADV to warn the operator.3. The computer-implemented method of claim 1 , wherein the threshold is approximately 2 seconds and the ...

Method and control unit for limiting an accident risk

A method for limiting an accident risk. The method initially includes a step of identifying, in which an increased accident risk is identified due to a changed situation in the interior of a vehicle and/or a changed traffic situation in the surroundings of the vehicle. The method also includes a step of providing, in which a control signal is provided for activating the vehicle to change a driving style and/or a travel route and/or an interior parameter in response to the identified change of the situation in the of the vehicle and/or traffic situation in the surroundings of the vehicle, in order to limit the accident risk. 113-. (canceled)14. A method for limiting an accident risk , the method comprising the following steps:identifying an increased accident risk due to a changed situation in an interior of a vehicle and/or a changed traffic situation in surroundings of the vehicle; andproviding, in response to the identifying of the increased accident risk due to the changed situation in the interior of the vehicle and/or the changed traffic situation in the surroundings of the vehicle, a control signal for controlling the vehicle to change a driving style of the vehicle, and/or a travel route of the vehicle, and/or an interior parameter of the vehicle, to limit an accident risk.15. The method as recited in claim 14 , wherein in the identifying step claim 14 , the changed situation in the vehicle interior is identified using: (i) a signal of an interior camera unit claim 14 , and/or (ii) a signal of a seat device claim 14 , the signal of the seat device representing a changed seat adjustment of the seat device.16. The method as recited in claim 15 , wherein in the identifying step claim 15 , a degree of an imminent injury severity of a vehicle occupant is determined using the changed situation in the interior of the vehicle.17. The method as recited in claim 14 , further comprising the following step:ascertaining a relative speed and/or a relative speed range, ...

GROUND TRUTH BASED METRICS FOR EVALUATION OF MACHINE LEARNING BASED MODELS FOR PREDICTING ATTRIBUTES OF TRAFFIC ENTITIES FOR NAVIGATING AUTONOMOUS VEHICLES

A system uses a machine learning based model to determine attributes describing states of mind and behavior of traffic entities in video frames captured by an autonomous vehicle. The system classifies video frames according to traffic scenarios depicted, where each scenario is associated with a filter based on vehicle attributes, traffic attributes, and road attributes. The system identifies a set of video frames associated with ground truth scenarios for validating the accuracy of the machine learning based model and predicts attributes of traffic entities in the video frames. The system analyzes video frames captured after the set of video frames to determine actual attributes of the traffic entities. Based on a comparison of the predicted attributes and actual attributes, the system determines a likelihood of the machine learning based model making accurate predictions and uses the likelihood to generate a navigation action table for controlling the autonomous vehicle. 1. A method comprising:identifying sets of video frames associated with ground truth scenarios, each set of video frames associated with an additional set of video frames captured within a threshold time interval of the set of video frames; providing the set of video frames as input to a machine learning based model that outputs one or more values associated with attributes describing a state of mind of a traffic entity in the set of video frames;', 'determining a predicted behavior of the traffic entity based on the one or more values;', 'determining an actual behavior of the traffic entity based on a corresponding additional set of video frames; and', 'comparing the predicted behavior and the actual behavior;, 'for each set of video frames associated with a ground truth scenariogenerating a navigation action table mapping output values of the machine learning based model to navigation actions to be performed by an autonomous vehicle based on comparisons of the predicted behavior and the actual ...

Vehicle control device

A vehicle control device according to an example in the present disclosure detects a monitoring target vehicle that may potentially cut into an own-lane from an adjacent lane. Further, the vehicle control device first executes an evasive preparation when sensing a predetermined relative motion of the monitoring target vehicle relative to the flow of the adjacent lane. The vehicle control device executes an evasive action to avoid interference between the monitoring target vehicle and the own-vehicle.

Systems and methods for long-term prediction of lane change maneuver

A method comprises making initial predictions of whether a first vehicle will perform a lane change at a plurality of future time steps based on sensor data captured by an egovehicle; and in response to making an initial prediction that the first vehicle will perform a lane change at a first one of the future time steps, making final predictions that the first vehicle will perform a lane change at each of a plurality of time steps subsequent to the first one of the future time steps.

Training a Generator Unit and a Discriminator Unit for Collision-Aware Trajectory Prediction

A system trains a generator unit and a discriminator unit simultaneously. The generator unit is configured to determine a future trajectory of at least one other road user in the environment of a vehicle considering an observed trajectory of the at least one other road user. The discriminator unit is configured to determine whether the determined future trajectory of the other road user is an actual future trajectory of the other road user. The system is configured to train the generator unit and the discriminator unit simultaneously with gradient descent. 1. A system , comprising:a generator unit and a discriminator unit,wherein said generator unit is configured to determine a future trajectory of at least one other road user in an environment of a vehicle considering an observed trajectory of the at least one other road user,wherein said discriminator unit is configured to determine whether the determined future trajectory of the at least one other road user is an actual future trajectory of the at least one other road user, andwherein said system is configured to train said generator unit and said discriminator unit simultaneously with gradient descent.2. The system according to claim 1 , wherein the other road user is a vulnerable road user.3. A system according to claim 1 , further comprising:an oracle unit,wherein said oracle unit is configured to determine a reward for the determined future trajectory of the at least one other road user considering whether the determined future trajectory of the other road user is collision-free, andwherein said system is configured to train said generator unit considering the reward determined by the oracle unit.4. The system according to claim 1 , whereinthe generator unit is configured to determine the future trajectory of the at least one other road user considering at least one static object in the environment of the other road user.5. The system according to claim 4 , whereinthe generator unit is configured to determine ...

DRIVING POLICIES DETERMINATION

A method for determining driving policies for a vehicle includes: receiving, in real-time during a trip of a vehicle, at least a set of input multimedia content elements captured by at least one sensor deployed in proximity to the vehicle, determining a plurality of possible future scenarios based at least on the set of input multimedia content elements, determining a probability score for each of the plurality of possible future scenarios, determining at least one driving policy according to at least the probability score for at least one of the plurality of possible future scenarios, and controlling the vehicle according to the at least one driving policy. 1. A method for determining driving policies for a vehicle , comprising:receiving, in real-time during a trip of a vehicle, at least a set of input multimedia content elements captured by at least one sensor deployed in proximity to the vehicle;determining, a plurality of possible future scenarios based at least on the set of input multimedia content elements;determining a probability score for each of the plurality of possible future scenarios;determining at least one driving policy according to at least the probability score for at least one of the plurality of possible future scenarios; andcontrolling the vehicle according to the at least one driving policy.2. The method of claim 1 , further comprising:generating, at least one signature for each one multimedia content element of the set of input multimedia content elements, wherein the plurality of possible future scenarios is determined by at least matching the at least one signature to at least one reference signature associated with at least one of the plurality of future scenarios.3. The method of claim 1 , wherein the set of multimedia content elements includes at least one of: an image; or a video.4. The method of claim 1 , wherein the set of multimedia content elements includes at least an audio signal.5. The method of claim 1 , wherein the determining ...

Vehicle Track Prediction Method and Device, Storage Medium and Terminal Device

A vehicle track prediction method and device, a storage medium and a terminal device are provided. The method includes: determining an obstacle vehicle entering a junction region in a case that an autonomous vehicle enters the junction region; acquiring historical traveling data of the obstacle vehicle in the junction region; predicting a potential track of the obstacle vehicle according to the historical traveling data and a current traveling state of the obstacle vehicle; and predicting a track of the autonomous vehicle in the junction region according to the potential track of the obstacle vehicle and a current traveling state of the autonomous vehicle. A decision making accuracy in self-driving may be effectively improved, and a driving risk may be reduced.

CONTROL METHOD AND CONTROL DEVICE FOR AUTONOMOUS VEHICLE

A control method for an autonomous vehicle is used in an autonomous vehicle including an engine, and a controller that controls an operation of the engine. In the control method, required driving force is set in accordance with an intervehicular distance between an own vehicle and a preceding vehicle when there is the preceding vehicle in front of the own vehicle. Also, when there is the preceding vehicle, a behavior of the preceding vehicle is predicted from a situation in front of the preceding vehicle. Further, when there is the preceding vehicle, sailing stop is executed based on the required driving force and the predicted behavior of the preceding vehicle. The sailing stop causes the engine to stop automatically while the own vehicle is traveling at vehicle speed equal to or higher than given vehicle speed. 18.-. (canceled)9. A control method for an autonomous vehicle provided with an engine as a driving source , wherein:setting a required driving force in accordance with an intervehicular distance between an own vehicle and a preceding vehicle when there is the preceding vehicle in front of the own vehicle;predicting a behavior of the preceding vehicle from a situation in front of the preceding vehicle when there is the preceding vehicle;executing a sailing stop based on the required driving force and the predicted behavior of the preceding vehicle, the sailing stop causing the engine to stop automatically while the own vehicle is traveling at vehicle speed equal to or higher than given vehicle speed; andwhen future deceleration of the preceding vehicle is predicted as the behavior of the preceding vehicle in response to expansion of the intervehicular distance, prohibiting a cancellation of the sailing stop for the engine that is automatically stopped.10. The control method for the autonomous vehicle according to claim 9 , wherein:detecting a traveling state of a pre-preceding vehicle traveling in front of the preceding vehicle, the traveling state being ...

Perception and Motion Prediction for Autonomous Devices

Systems, methods, tangible non-transitory computer-readable media, and devices associated with object perception and prediction of object motion are provided. For example, a plurality of temporal instance representations can be generated. Each temporal instance representation can be associated with differences in the appearance and motion of objects over past time intervals. Past paths and candidate paths of a set of objects can be determined based on the temporal instance representations and current detections of objects. Predicted paths of the set of objects using a machine-learned model trained that uses the past paths and candidate paths to determine the predicted paths. Past path data that includes information associated with the predicted paths can be generated for each object of the set of objects respectively. 1. A computer-implemented method of perception and motion forecasting , the computer-implemented method comprising:generating, by a computing system comprising one or more computing devices, a plurality of temporal instance representations, wherein each temporal instance representation is associated with differences in an appearance and a motion of one or more objects over past time intervals;determining, by the computing system, based at least in part on the plurality of temporal instance representations and current detections of a set of objects comprising the one or more objects, one or more past paths of the one or more objects over the past time intervals and one or more candidate paths of the set of objects over a set of time intervals comprising a current time interval and at least one of the past time intervals;determining, by the computing system, one or more predicted paths of the set of objects based at least in part on one or more machine-learned models, the one or more machine-learned models utilizing the one or more past paths and the one or more candidate paths to infer the one or more predicted paths; andgenerating, by the computing ...

VEHICLE TO VEHICLE (V2V) COMMUNICATION LESS TRUCK PLATOONING

A method for at least partial autonomous driving based on one or more other vehicles, the method may include (i) detecting a trusted vehicle within an environment of a first vehicle, the first vehicle is destined to follow a first path and reach a first target location, the trusted vehicle is destined to follow a second path; (ii) determining, based on at least a spatial relationship between the first path and the second path, at least one out of: (a) whether to perform an at least partial autonomous driving session that involves at least partially automatically following the trusted vehicle, and (b) whether to suggest to a human driver of the first vehicle to authorize the first vehicle to perform the at least partial autonomous driving session; and responding to the determination. 1. A method for at least partial autonomous driving based on one or more other vehicles , the method comprises:detecting a trusted vehicle within an environment of a first vehicle, the first vehicle is destined to follow a first path and reach a first target location, the trusted vehicle is destined to follow a second path; (a) whether to perform an at least partial autonomous driving session that involves at least partially automatically following the trusted vehicle, and', '(b) whether to suggest to a human driver of the first vehicle to authorize the first vehicle to perform the at least partial autonomous driving session; and, 'determining, based on at least a spatial relationship between the first path and the second path, at least one out ofresponding to the determination;wherein the responding comprises at least one out of:(a) performing the at least partial autonomous driving session when determining to perform the at least partial autonomous driving session; and(b) suggesting the human driver to authorize the first vehicle to perform the at least partial autonomous driving session, when determining to suggest the human driver to authorize the first vehicle to perform the at ...

Autonomous Vehicle Operational Management Scenarios

Traversing, by an autonomous vehicle, a vehicle transportation network, may include operating a scenario-specific operational control evaluation module instance, wherein the scenario-specific operational control evaluation module instance includes an instance of a scenario-specific operational control evaluation model of a vehicle operational scenario wherein the vehicle operational scenario is a merge vehicle operational scenario or a pass-obstruction vehicle operational scenario, receiving a candidate vehicle control action from the scenario-specific operational control evaluation module instance, and traversing a portion of the vehicle transportation network in accordance with the candidate vehicle control action. 1. A method for use in traversing a vehicle transportation network , the method comprising: operating a scenario-specific operational control evaluation module instance, wherein the scenario-specific operational control evaluation module instance includes an instance of a scenario-specific operational control evaluation model of a vehicle operational scenario wherein the vehicle operational scenario is a merge vehicle operational scenario or a pass-obstruction vehicle operational scenario, wherein operating the scenario-specific operational control evaluation module instance includes identifying a policy for the scenario-specific operational control evaluation model;', 'receiving a candidate vehicle control action from the scenario-specific operational control evaluation module instance; and', 'traversing a portion of the vehicle transportation network in accordance with the candidate vehicle control action., 'traversing, by an autonomous vehicle, a vehicle transportation network, wherein traversing the vehicle transportation network includes2. The method of claim 1 , wherein traversing the vehicle transportation network includes:in response to receiving, from an operational environment monitor of the vehicle, operational environment information ...

Vehicle control device

In a vehicle control device, a target motion estimation unit estimates a motion for a target area representing an area where no other vehicle is traveling on a lane in the future based on motions of other vehicles traveling on the lane to which a course is to be changed. A control amount setting unit sets a control amount of an own vehicle required in order to make the motion of the target area and a motion of the own vehicle match. A motion control unit controls the motion of the own vehicle according to the set control amount.

METHODS AND SYSTEMS FOR COMPUTER-BASED DETERMINING OF PRESENCE OF DYNAMIC OBJECTS

A method for determining a set of dynamic objects in sensor data representative of a surrounding area of a vehicle having sensors, the method being executed by a server, the server executing a machine learning algorithm (MLA). Sensor data is received, and the MLA generates, based on the sensor data, a set of feature vectors. Vehicle data indicative of a localization of the vehicle is received. The MLA generates, based on the set of feature vectors and the vehicle data, a tensor, the tensor including a grid representation of the surrounding area. The MLA generates an mobility mask indicative of grid cells occupied by at least one moving potential object in the grid, and a velocity mask indicative of a velocity associated with the at least one potential object in the grid. The MLA determines, based on the mobility mask and the velocity mask, the set of dynamic objects. 1. A computer-implemented method for determining a set of dynamic objects based on sensor data acquired by a sensor mounted on a vehicle , the method being executable by an electronic device , the electronic device communicatively coupled to the sensor for collection of sensor data therefrom , the electronic device executing a machine learning algorithm (MLA) having been trained for object detection based on sensor data , the method comprising:receiving sensor data representative of a surrounding area of the vehicle;generating, by the MLA, based on the sensor data, a set of feature vectors of the surrounding area;receiving vehicle data indicative of a localization of the vehicle on a map;generating, by the MLA, based on the set of feature vectors and the vehicle data, a tensor, the tensor including a grid representation of the surrounding area; an mobility mask indicative of grid cells occupied by at least one potential object in the grid, and', 'a velocity mask indicative of a velocity associated with the at least one potential object in the grid; and, 'generating, by the MLA, based on the ...

Method for the Autonomous Driving of a Vehicle in a Narrow Passage

The invention relates to a method for the autonomous driving of a vehicle ( 1 ) through a narrow passage ( 5 ). The method comprises the storing of a right-of-way rule for the narrow passage ( 5 ) in a database, which the vehicle ( 1 ) can access. The narrow passage ( 5 ) and an oncoming vehicle ( 13 ) approaching in the region of the narrow passage ( 5 ) are sensed by means of sensors of the vehicle, and the speed of the oncoming vehicle ( 13 ) is determined by means of speed data sensed by the sensors. Furthermore, a reaction of the oncoming vehicle ( 13 ) is predicted in accordance with the determined speed of the oncoming vehicle ( 13 ), and the vehicle ( 1 ) is moved through the narrow passage ( 5 ), provided that the prediction of the reaction of the oncoming vehicle ( 13 ) indicates that the oncoming vehicle ( 13 ) will not pass through the narrow passage or frees the narrow passage ( 5 ) for passage or the right-of-way rule stored in the database provides that the vehicle ( 1 ) has the right of way in the narrow passage ( 13 ).

Vehicle driving control apparatus and method

본 발명은 차량 외부에 위치하는 오브젝트를 감지하는 오브젝트 감지부; 및 상기 오브젝트에 대한 정보에 기초하여, 제1 시간 범위 동안, 차량의 주행 속도가, 증가하거나 감소되도록 제어하는 제1 제어 신호를 제공하고, 제2 시간 범위 동안, 상기 제1 제어 신호에 따른 차량의 제어와 반대되게 차량이 제어되도록 제2 제어 신호를 제공하는 프로세서;를 포함하는 차량 주행 제어 장치에 관한 것이다. The present invention provides an object detecting unit for detecting an object located outside the vehicle; And a first control signal for controlling the traveling speed of the vehicle to increase or decrease during the first time range based on the information on the object, and during the second time range, the vehicle according to the first control signal. And a processor configured to provide a second control signal to control the vehicle as opposed to control of the vehicle.

用于导航车辆的系统和方法

一种自主系统可以选择性地取代对主车辆的人类驾驶员控制。该系统可以接收代表主车辆环境的图像，并基于对图像的分析来检测主车辆环境中的障碍物。该系统可以监视对与主车辆相关联的油门、制动或转向控件的驾驶员输入。该系统可以确定驾驶员输入是否会导致主车辆在相对于障碍物的接近缓冲区内导航。如果驾驶员输入不会导致主车辆在接近缓冲区内导航，则该系统可以允许驾驶员输入引起一个或多个主车辆运动控制系统中的对应改变。如果驾驶员输入会导致主车辆在接近缓冲区内导航，则该系统可以阻止驾驶员输入引起对应改变。

Systems and methods for navigating a vehicle

자율 시스템은 호스트 차량의 인간 운전자 제어를 선택적으로 배제할 수 있다. 상기 시스템은 상기 호스트 차량의 주변상황을 나타내는 이미지를 수신하고 상기 이미지의 분석에 의거하여 상기 호스트 차량의 상기 주변상황에서 장애물을 검출할 수 있다. 상기 시스템은 상기 호스트 차량과 연관된 구동, 제동, 및/또는 조향 제어로의 운전자 입력을 모니터링 할 수 있다. 상기 시스템은 상기 운전자 입력의 결과로 상기 호스트 차량이 상기 장애물에 대한 근접 버퍼 이내로 주행하게 되는지 여부를 판단할 수 있다. 운전자 입력의 결과고 상기 호스트 차량이 상기 근접 버퍼 이내로 주행하지 않게 될 경우, 상기 시스템은 운전자 입력이 하나 이상의 호스트 차량 동작 제어 시스템에 상응하는 변경을 유발하도록 허용할 수 있다. 상기 운전자 입력의 결과로 상기 호스트 차량이 상기 근접 버퍼 이내로 주행하게 될 경우, 상기 시스템은 상기 운전자 입력이 상기 상응하는 변경을 유발하는 것을 방지할 수 있다. The autonomous system may selectively exclude human driver control of the host vehicle. The system may receive an image representing the surrounding situation of the host vehicle and detect an obstacle in the surrounding situation of the host vehicle based on the analysis of the image. The system may monitor driver inputs to drive, brake, and/or steering controls associated with the host vehicle. The system may determine whether the host vehicle is driven within a proximity buffer for the obstacle as a result of the driver input. If, as a result of driver input, the host vehicle does not travel within the proximity buffer, the system may allow the driver input to cause a corresponding change in one or more host vehicle motion control systems. If, as a result of the driver input, the host vehicle is driven into the proximity buffer, the system may prevent the driver input from causing the corresponding change.

Systems and methods for navigation of vehicles

자율 시스템은 호스트 차량의 인간 운전자 제어를 선택적으로 배제할 수 있다. 상기 시스템은 상기 호스트 차량의 주변상황을 나타내는 이미지를 수신하고 상기 이미지의 분석에 의거하여 상기 호스트 차량의 상기 주변상황에서 장애물을 검출할 수 있다. 상기 시스템은 상기 호스트 차량과 연관된 구동, 제동, 및/또는 조향 제어로의 운전자 입력을 모니터링 할 수 있다. 상기 시스템은 상기 운전자 입력의 결과로 상기 호스트 차량이 상기 장애물에 대한 근접 버퍼 이내로 주행하게 되는지 여부를 판단할 수 있다. 운전자 입력의 결과고 상기 호스트 차량이 상기 근접 버퍼 이내로 주행하지 않게 될 경우, 상기 시스템은 운전자 입력이 하나 이상의 호스트 차량 동작 제어 시스템에 상응하는 변경을 유발하도록 허용할 수 있다. 상기 운전자 입력의 결과로 상기 호스트 차량이 상기 근접 버퍼 이내로 주행하게 될 경우, 상기 시스템은 상기 운전자 입력이 상기 상응하는 변경을 유발하는 것을 방지할 수 있다. The autonomous system can selectively exclude human driver control of the host vehicle. The system may receive an image representing the surrounding situation of the host vehicle and detect an obstacle in the surrounding situation of the host vehicle based on the analysis of the image. The system can monitor driver input to drive, braking, and/or steering control associated with the host vehicle. The system may determine whether the host vehicle is driven within a proximity buffer for the obstacle as a result of the driver input. If it is the result of driver input and the host vehicle will not drive within the proximity buffer, the system may allow the driver input to cause a change corresponding to one or more host vehicle motion control systems. When the host vehicle is driven within the proximity buffer as a result of the driver input, the system can prevent the driver input from causing the corresponding change.

Object tracking supporting autonomous vehicle navigation

본 개시는 일반적으로 자율 주행 차량에 근접한 대상체를 광학적으로 추적하기 위한 시스템 및 방법에 관한 것이다. 상세하게는, 대상체 추적 시스템은 대상체의 이전에 결정된 위치에 적어도 부분적으로 기초하여 자율 주행 차량 주위의 대상체의 위치를 결정하는 것에 의해 추적되는 대상체에 대한 위치 데이터를 세분화할 수 있다. 특정 경우에, 대상체에 대한 세분화된 위치에 도착하는 데 사용되는 예측된 위치 및 검출된 위치는 센서 데이터의 조건 및 과거 데이터의 품질에 따라 상이한 방식으로 가중될 수 있다.

Trajectory planner

本发明提供一种能够进行轨线预测的自动汽车，该自动汽车可以包括第一传感器、第二传感器、处理器、轨线规划器、低水平控制器和汽车致动器。该第一传感器可以是第一传感器类型，并且可以检测障碍物和目标。该第二传感器可以是第二传感器类型，并且可以检测该障碍物和该目标。该处理器可以对由该第一传感器检测到的该障碍物和由该第二传感器检测到的该障碍物执行匹配，基于该匹配对该障碍物的存在概率进行建模，以及基于该存在概率和恒定速度模型来追踪该障碍物。该轨线规划器可以基于追踪的障碍物、该目标和非线性模型预测控制(NMPC)来为该自动汽车生成轨线。该低水平控制器可以通过驱动汽车致动器来为该自动汽车实现该轨线。

Action planning system and method for the autonomous vehicles

提供了用于自主交通工具的动作规划系统（100）和方法。系统（100）包括：一个或多个处理器（108）；和一个或多个非暂时性计算机可读存储介质（110），其上存储有由一个或多个处理器（108）使用的计算机程序，其中计算机程序引起一个或多个处理器（108）估计自主交通工具（114）的未来环境，为自主交通工具（114）生成可能的轨迹，基于当前的本地交通情境预测自主交通工具（114）的未来环境中的每个动态障碍物的运动和反应，以及在时间步内迭代地生成预测。

Autonomous driving system for preventing collision of cut-in vehicle and autonomous driving method thereof

본 발명의 일 실시예에 따른 컷인 차량의 충돌 방지를 위한 자율주행시스템은 주변차량의 주행 정보를 검출하여 차량제어부로 전달하는 상대차검출부, 자차의 주행 정보를 검출하여 차량제어부로 전달하는 자차검출부 및 상기 자차의 전방으로 진입하는 주변차량과 충돌을 피하기 위한 복수의 회피경로를 생성하고, 상기 회피경로 상에 일정 간격으로 속도에 따른 도달위치를 지정하며, 상기 주변차량과 상기 자차의 충돌 가능성이 존재하면, 상기 복수의 회피경로 중 충돌을 피할 수 있는 회피경로를 선택한 후 상기 선택한 회피경로 상에서 상기 주변차량과 충돌을 피할 수 있는 도달위치에 도착할 수 있도록 상기 자차의 속도를 제어하는 차량제어부를 포함할 수 있다. An autonomous driving system for preventing collision of a cut-in vehicle according to an embodiment of the present invention includes a counterpart vehicle detection unit that detects driving information of a surrounding vehicle and transmits it to the vehicle control unit, a vehicle detection unit that detects driving information of the own vehicle and transmits it to the vehicle control unit, and Creates a plurality of avoidance routes to avoid collision with neighboring vehicles entering the front of the own vehicle, designates arrival positions according to speed at regular intervals on the avoidance route, and there is a possibility of collision between the surrounding vehicle and the own vehicle Then, after selecting an avoidance route capable of avoiding a collision among the plurality of avoidance routes, a vehicle control unit controlling the speed of the own vehicle so as to arrive at a location where a collision with the surrounding vehicle can be avoided on the selected avoidance route. can

Trajectory classification

本文讨论了预测环境中对象行为的技术。例如，这种技术可以包括将数据输入到模型中，并从模型中接收代表离散化表示的输出。该离散化表示可与对象在未来时间到达环境中一位置的概率相关联。车辆计算系统可以使用离散化表示和概率来确定轨迹和与轨迹相关联的权重。车辆(比如自主车辆)可以基于车辆计算系统输出的轨迹和权重而受到控制，以穿越环境。

Driving support method and driving support device

Control method and control device for automatic driving vehicle

自动驾驶车辆的控制方法在具有发动机以及控制发动机的动作的控制器的自动驾驶车辆中，当在本车辆前方存在在前车辆的情况下，设定与本车辆和在前车辆之间的车间距离相应的请求驱动力，在存在在前车辆的情况下，根据在前车辆前方的状况预测该在前车辆的行动。并且，在存在在前车辆的情况下，基于请求驱动力以及预测出的在前车辆的行动，实施在以大于或等于规定车速的车速行驶中使发动机自动地停止的巡航停止。

Vehicle trajectory modification for following

本发明讨论了用于确定基于物体来修改轨迹的技术。车辆可以确定环境的可行驶区域，捕获代表该环境中的物体的传感器数据，并执行抽查以确定是否修改轨迹。这种抽查可以包括将物体的实际或预测范围合并到可行驶区域中以修改可行驶区域的处理。可以在沿着参考轨迹的离散点处确定参考轨迹与物体之间的距离，并且基于与轨迹和修改区域相关联的成本，距离或相交，车辆可以修改其轨迹。一种轨迹修改包括以下内容，其可包括改变车辆的纵向控制，例如以维持车辆与物体之间的相对距离和速度。

Method, device, computing equipment and storage medium for predicting lane change of vehicle

本申请公开了一种预测车辆换道的方法，包括：获取拍摄图像，该拍摄图像包括目标车辆以及车道线分别对应的成像；根据拍摄图像，确定目标车辆的横向移动方向、目标车辆与本车之间的横向距离以及目标车辆的换道概率，从而根据该横向移动方向、横向距离以及换道概率，预测目标车辆是否进行换道。其中，目标车辆的换道概率通过目标车辆与目标车道线之间的相对位置进行确定，该目标车道线为在横向移动方向上距离目标车辆最近的车道线。由于上述预测车辆是否进行换道的过程中，不受车辆类型差异的影响，从而可以提高车辆行驶的安全性。此外，本申请还提供了对应的装置、计算设备及存储介质。

Method and system for predicting lane changing intention of front vehicle

本发明属于自动驾驶技术领域，公开了一种前车换道意图预测方法及预测系统，前车换道意图预测方法包括：建立以自车车头中心点为零点的动态栅格地图，基于语义分割的方法获取前方车辆的轮廓特征与车道线，并检测前方障碍车辆车头中心与自车的相对距离，用栅格坐标表示前方障碍车辆实时位置，进行前车的位置信息的描述，并根据前车的横纵向速度，分析前方障碍车辆的相关运动状态信息，将动态栅格地图与混合高斯隐马尔科夫模型相结合，进行前车换道行为的预测。本发明提出的基于动态栅格地图与混合高斯隐马尔科夫模型的前车换道意图预测方法可以达到较好的预测效果。本发明可利用可观测到的前车的行驶状态，预测其隐藏的换道行为状态，为自车安全行驶提供相关信息。

Apparatus and method for autonomous driving

본 발명의 일 실시예에 따른 자율주행장치는 주변 차량의 주행 정보를 검출하여 차량제어부로 전달하는 상대차검출부, 자차의 주행 정보를 검출하여 차량제어부로 전달하는 자차검출부 및 상기 자차의 전방으로 진입하는 주변 차량과 충돌을 피하기 위한 복수의 회피 경로를 생성하고, 상기 회피 경로 상에 일정 간격으로 속도에 따른 도달 위치를 지정하며, 상기 주변 차량의 속도 또는 회전 방향을 기초로 상기 자차의 충돌 가능성이 존재할 경우 상기 복수의 회피 경로 중 최소 거리를 이동하면서 상기 주변 차량과 충돌을 피할 수 있는 회피 경로를 선택한 후 상기 선택한 회피 경로 상에서 상기 주변 차량과 충돌을 피할 수 있는 도달 위치에 상기 자차가 도착할 수 있도록 상기 자차의 속도를 제어하는 것으로 상기 주변 차량의 주행 상태 및 상기 자차와 상기 주변 차량의 충돌 형태에 따라 상기 자차의 도달 위치가 정해지도록 자차의 속도를 제어하는 차량제어부를 포함할 수 있다.

System and method for predicting the trajectory of an ego vehicle as a function of the environment of said ego vehicle

Procédé (40) de prédiction de la trajectoire d’un véhicule égo dans lequel : - on récupère des données provenant de capteurs extéroceptifs et proprioceptifs ; - on modélise un environnement du véhicule égo en fonction d’un ensemble desdites données récupérées ; - on prédit la trajectoire du véhicule égo et au moins une trajectoire d’objet mobile situé dans l’environnement modélisé du véhicule égo en générant une trajectoire prédite du véhicule égo pendant une durée déterminée en prenant en compte l’environnement modélisé du véhicule ; - on sélectionne un objet cible ; et - on transmet la valeur de prédiction de trajectoire à l’unité de contrôle du véhicule automobile égo. Figure pour l’abrégé : Fig 3

Method and device for determining an acceleration of a moving object in an environment of a vehicle

L’invention concerne un procédé et un dispositif de détermination de l’accélération d’un objet mobile se déplaçant dans l’environnement d’un véhicule. A cet effet, l’accélération courante de l’objet mobile est déterminée à un instant courant à partir de données obtenues de capteurs de détection d’objet embarqués dans le véhicule. Une valeur représentative d’une variation d’accélération, aussi appelée secousse (ou « jerk » en anglais) et associée à l’objet mobile est obtenue. Une telle valeur de secousse est supérieure ou égale à 0 et correspond à une valeur constante. L’évolution (2) de l’accélération de l’objet mobile en fonction du temps est déterminée en fonction de l’accélération courante et de la valeur de secousse associée à l’objet mobile de telle manière que l’accélération de l’objet mobile tende vers 0 après écoulement d’une durée déterminée. Figure pour l’abrégé : Figure 2

Apparatus for Controlling Vehicle, System Including Same and Method Thereof

An apparatus for controlling a vehicle includes an object selection device configured to select an object intersecting the vehicle at an intersection existing on a driving path of the vehicle, a risk determination device configured to determine a risk during driving of the vehicle based on a predicted path of the object, and a driving control device configured to determine a driving method of the vehicle based on a risk determination result.

Vehicle travel control device and method

本发明提供一种车辆行驶控制装置，包括：对象检测部，检测位于车辆的外部的对象；以及处理器，基于所述对象相关的信息来按第一时间范围期间提供用于控制为使车辆的行驶速度增大或减小的第一控制信号，并按第二时间范围期间提供用于以与基于所述第一控制信号的车辆的控制相反的方式控制所述车辆的第二控制信号。

Driving assistance apparatus

A driving assistance apparatus includes a host vehicle information acquisition device configured to acquire host vehicle information including a vehicle speed of a host vehicle and a signal indicating that braking force is applied to the host vehicle by a braking device of the host vehicle, a target information acquisition device configured to acquire target information including a relative position of a target present on the periphery of the host vehicle for the host vehicle, an advancing direction of the target, and a speed of the target, and an electronic control device.

Systems and methods for navigating a vehicle

An autonomous system may selectively displace human driver control of a host vehicle. The system may receive an image representative of an environment of the host vehicle and detect an obstacle in the environment of the host vehicle based on analysis of the image. The system may monitor a driver input to a throttle, brake, and/or steering control associated with the host vehicle. The system may determine whether the driver input would result in the host vehicle navigating within a proximity buffer relative to the obstacle. If the driver input would not result in the host vehicle navigating within the proximity buffer, the system may allow the driver input to cause a corresponding change in one or more host vehicle motion control systems. If the driver input would result in the host vehicle navigating within the proximity buffer, the system may prevent the driver input from causing the corresponding change.

Tracking vanished objects for autonomous vehicles

Driving assistance method and driving assistance device

A driving assistance method includes: detecting (S4) a first other vehicle (v1) entering an intersection on a first route (21) where a host vehicle (20) is traveling from a second route (21) different from the first route (21); predicting (S7) whether or not the first other vehicle (v1) will stop in the intersection, and predicting a stop position of the first other vehicle (v1) when the first other vehicle (v1) is predicted to stop in the intersection; calculating (S8) a minimum distance (d1) of a first gap between a vehicle body of the first other vehicle (v1) and a surrounding object around the first other vehicle (v1) or between the vehicle body of the first other vehicle (v1) and a road edge of a travel lane of the first other vehicle (v1) when the first other vehicle (v1) stops at the predicted stop position; and predicting (S9, S12, S18) according to the calculated minimum distance (d1) whether or not a second other vehicle (v2), which is a following vehicle behind the first other vehicle (v1), may slip through the first gap from behind the first other vehicle (v1).

Method and device for autonomous driving of a land motor vehicle

L’invention concerne un procédé de conduite autonome d’un véhicule terrestre à moteur, appelé égo-véhicule 102, par régulation de vitesse adaptative fondée sur une vitesse cible. Le procédé est apte à détecter un premier véhicule cible 104 et à calculer un indicateur de pertinence pour le premier véhicule cible 104 configuré pour caractériser une probabilité de présence du premier véhicule cible. Sur détection d’une perte de détection du premier véhicule cible 104 et si l’indicateur de pertinence est supérieur à une valeur de désélection prédéterminée, le procédé calcule d’une consigne de conduite autonome à partir de l’information de mouvement du premier véhicule cible 104, de la vitesse cible et de la vitesse de l’égo-véhicule 102. Figure pour l’abrégé: Figure 1 The invention relates to a method for autonomous driving of a motorized land vehicle, called ego-vehicle 102, by adaptive cruise control based on a target speed. The method is able to detect a first target vehicle 104 and to calculate a relevance indicator for the first target vehicle 104 configured to characterize a probability of presence of the first target vehicle. Upon detection of a loss of detection of the first target vehicle 104 and if the relevance indicator is greater than a predetermined deselection value, the method calculates an autonomous driving instruction from the movement information of the first vehicle target 104, target speed and ego-vehicle speed 102. Figure for the abstract: Figure 1

Intelligent vehicle forced lane change merge point determination method and device

本发明公开了一种智能车强制换道汇入点确定方法及装置，方法包括：获取目标车辆换道微观信息，包括目标车辆、目标车道内前导车辆和跟随车辆的速度、位置数据；计算目标车辆和目标车道内跟随车辆的临界安全间隙；根据自由换道和强制换道类型，计算有效换道数量的礼貌因子均值；确定代表驾驶特性的权重系数，计算当前智能车强制换道的礼貌因子；根据目标车辆和目标车道内前导车辆的位置关系，计算确定当前智能车强制换道的汇入点。本发明提供的方法综合考虑换道车辆与目标车道内前后车的相互作用，换道汇入点选择更加精确，进而为驾驶员或智能车提供科学合理的判断和决策依据，为道路交通安全和行驶效率提供保障。

Unprotected crossroad unmanned vehicle rolling optimization decision method

本发明提供了一种无保护十字路口无人车滚动优化决策方法，获取本车和障碍物的位置及速度信息以及路口的车道线信息，生成本车的参考轨迹及预测周车轨迹；然后根据本车的参考轨迹对车辆信息进行Frenet坐标转换，得到Frenet坐标系下的本车和周车轨迹；接着建立Frenet坐标系下的无人车十字路口决策模型，来描述本车在无保护十字路口的运动；最后设计模型预测控制的无保护十字路口决策控制器，通过求解优化问题并将纵向速度和期望轨迹作用到底层控制器即可实现基于模型预测控制的无保护十字路口无人车行为决策；本方法在无保护十字路口无人车行为决策过程中将本车与周围物体的相对位置关系在纵向和横向上分离开，更能描述本车与周车的碰撞危险。

Vehicle trajectory modification for following

Techniques for determining to modify a trajectory based on an object are discussed herein. A vehicle can determine a drivable area of an environment, capture sensor data representing an object in the environment, and perform a spot check to determine whether or not to modify a trajectory. Such a spot check may include processing to incorporate an actual or predicted extent of the object into the drivable area to modify the drivable area. A distance between a reference trajectory and the object can be determined at discrete points along the reference trajectory, and based on a cost, distance, or intersection associated with the trajectory and the modified area, the vehicle can modify its trajectory. One trajectory modification includes following, which may include varying a longitudinal control of the vehicle, for example, to maintain a relative distance and velocity between the vehicle and the object.

Lane detection method and system for a vehicle

The present disclosure is directed toward a lane detection method. The method includes: acquiring road information for a section of a road upon which a subject vehicle is traveling, determining whether an adjacent vehicle is traveling in vicinity of the subject vehicle, defining, in response to determining the presence of the adjacent vehicle, a vehicle trajectory based on movement of the adjacent vehicle and the road information, detecting one or more lane markings along the road upon which the subject vehicle is traveling, defining a lane marking trajectory in response to detecting the one or more lane markings, calculating a lane accuracy for each estimated lane trajectory that includes the vehicle trajectory, the lane marking trajectory, or a combination thereof, and selecting a drive lane from among the estimated lane trajectories based on the lane accuracy.

Automatic driving control device and automatic driving control method

本发明涉及自动驾驶控制装置和自动驾驶控制方法，所述方法包括收集自动行驶的本车和至少一个其他车辆的行驶信息，基于其他车辆的行驶信息确定其他车辆的行驶意图，基于其他车辆的行驶意图预测其他车辆的行驶路线，并且基于预测出的其他车辆的行驶路线确定本车的行驶路线。

Automatic driving vehicle and dynamic planning method and system for motion trail of automatic driving vehicle

本发明提供了一种自动驾驶车辆运动轨迹的动态规划方法，动态规划方法包括：获取当前时刻自动驾驶车辆的位置；感知自动驾驶车辆周围环境的环境数据；从环境数据中提取关于车道的车道信息；根据自动驾驶车辆当前时刻的位置、高清地图和车道信息获取自动驾驶车辆的第一可行驶区域；从环境数据中提取关于静态物体的静态信息；从环境数据中提取关于动态物体的动态信息，并根据动态信息预测动态物体的运动轨迹；根据第一可行驶区域、静态信息、运动轨迹、以及车道信息规划出第二可行驶区域。此外，本发明还提供了一种自动驾驶车辆及运动轨迹的动态规划系统。本发明技术方案有效解决了如何合理规划自动驾驶车辆运动轨迹的问题。

Methods and systems for determining the presence of dynamic objects by a computer

FIELD: computer engineering. SUBSTANCE: disclosed is a computer-based method of determining a set of dynamic objects based on sensor data obtained by a sensor mounted on a vehicle, performed by an electronic device, connected to the sensor to receive sensor data therefrom and executing a machine learning algorithm (MLA), trained to detect objects based on the sensor data, and including: receiving sensor data on the surrounding area of the vehicle; formation by the MLA algorithm based on the sensor data a set of the feature vectors of the surrounding area; receiving vehicle data indicating the location of the vehicle on the map; generation of tensor by MLA algorithm based on set of feature vectors and vehicle data, including a grid representation of the surrounding area; formation by the MLA algorithm based on the tensor of the mobility mask indicating the grid cells occupied by at least one potential object in the grid, and a velocity mask indicative of the velocity associated with at least one potential object in the grid; and determining, by the MLA algorithm, based on the mobility mask and the speed mask of the set of dynamic objects in the surrounding area of the vehicle. EFFECT: providing determination of a set of dynamic objects based on sensor data obtained by a sensor installed on a vehicle. 19 cl, 8 dwg РОССИЙСКАЯ ФЕДЕРАЦИЯ (19) RU (11) (13) 2 767 955 C1 (51) МПК B60W 30/08 (2012.01) G06N 20/00 (2019.01) ФЕДЕРАЛЬНАЯ СЛУЖБА ПО ИНТЕЛЛЕКТУАЛЬНОЙ СОБСТВЕННОСТИ (12) ОПИСАНИЕ ИЗОБРЕТЕНИЯ К ПАТЕНТУ (52) СПК B60W 30/08 (2021.08); G06N 20/00 (2021.08) (21)(22) Заявка: 2019116279, 27.05.2019 (24) Дата начала отсчета срока действия патента: Дата регистрации: (73) Патентообладатель(и): Общество с ограниченной ответственностью "Яндекс Беспилотные Технологии" (RU) 22.03.2022 (45) Опубликовано: 22.03.2022 Бюл. № 9 2 7 6 7 9 5 5 R U (54) СПОСОБЫ И СИСТЕМЫ ДЛЯ ОПРЕДЕЛЕНИЯ КОМПЬЮТЕРОМ НАЛИЧИЯ ДИНАМИЧЕСКИХ ОБЪЕКТОВ (57) Реферат: Изобретение относится к области набора векторов ...

Autonomous Driving Control Apparatus and Method

An autonomous driving control method includes collecting travel information on a host vehicle traveling autonomously and on at least one other vehicle, determining a driving intention of the other vehicle based on the travel information on the other vehicle, predicting a driving route of the other vehicle based on the driving intention of the other vehicle, and determining a driving route of the host vehicle based on the predicted driving route of the other vehicle.