Trajectory generation using road network model

A method and system of trajectory generation using a road network model comprising; obtaining, using at least one processor of a vehicle, a location of the vehicle, sensor data collected at the location; obtaining and map data for the location (701). Generating, using one or more processors, at least one possible trajectory for at least one object at the location, wherein the possible trajectory is constrained in accordance with the map data (702). Predicting, using a machine learning model, a score for the at least one trajectory (703). The system may generate feature vectors generated from an image embedding of the sensor data and the map data. Point clouds may be utilised in conjunction with LiDAR.

Passenger health screening and monitoring

Method of determining health conditions of vehicle occupants in a first trip from sensor 1008, and adjusting a second trip of the vehicle based on the condition of the occupants of the first trip (see para 119). It may include carrying few users or only users at low risk of infection during the following journey (next car share/car pool trip) due to the heath condition of the user from the previous journey. Risk of infection of the user of the following journey may be determined based on the subsequent users’ demographics. The route of the second trip may be adjusted such as to include few users or only those who are at low risk of infection; and the interior of the vehicle may be modified (e.g. seating arrangement, partition, airflow circulation, see para 114-116).

Conditional motion predictions

Conditional motion module 1302 of host vehicle 1) generates a set of candidate trajectories for a manoeuvre, then 2) predicts response of nearby vehicles to the trajectories, and the probability of the response, and finally 3) the ideal trajectory is selected based on the predicted responses 1500 and the characteristics of the trajectories. The manoeuvre action may be: a merge, lane change, a turn, lane follow at a determined speed, or lane change with a time limit. The response of nearby vehicles may be predicted base on their current tracked trajectory history. The selection of trajectory may involve a score of: magnitude of acceleration, potential of collision, time to collision, distance to collision, or distance travelled.

Passenger health screening and monitoring

Passenger health screening and monitoring

Method of determining health conditions of vehicle occupants from sensor data (or historic sensor data profile of occupants), the vehicle responds to detected medical conditions. Sensors may detect a cough, respiratory rate, motion, facial features, skin tone, facial expression, or body temperature of the occupant, and the pathogen/odour such as bacteria, virus, allergens, dust, within the vehicle. Mitigation action may include altering the trajectory or destination, rerouting; identifying and navigating to a safe stopping location to park, or to an emergency services location (hospital, police station, clinic, fire station); altering the driving style (acceleration, deceleration, braking); adjusting a seat of the user; engaging a partition screen (blinds, curtains, dividers, shutters); altering air flow, climate control or temperature; dispensing first aid, medication; alerting the user, or emergency services of the vehicle’s location or health conditions; dispensing disinfectant; operate ...

Road surface condition guided decision making and prediction

Passenger health screening and monitoring

Method of determining health conditions of vehicle occupants from sensor data (or historic sensor data profile of occupants), the vehicle responds to detected medical conditions. Sensors may detect a cough, respiratory rate, motion, facial features, skin tone, facial expression, or body temperature of the occupant, and the pathogen/odour such as bacteria, virus, allergens, dust, within the vehicle. Mitigation action may include altering the trajectory or destination, rerouting; identifying and navigating to a safe stopping location to park, or to an emergency services location (hospital, police station, clinic, fire station); altering the driving style (acceleration, deceleration, braking); adjusting a seat of the user; engaging a partition screen (blinds, curtains, dividers, shutters); altering air flow, climate control or temperature; dispensing first aid, medication; alerting the user, or emergency services of the vehicle’s location or health conditions; dispensing disinfectant; operate ...

Road surface condition guided decision making and prediction

Road surface condition guided decision making and prediction

A surface detection system for an autonomous vehicle comprising at least one sensor, a computer readable medium and at least one processor. The system receives data from the sensor associated with a surface along which the vehicle is travelling (1402). Using a surface classifier to determine a classification of the surface based on the sensor data (1404). Based on this data determining the drivability of the surface (1406) and planning based on the drivability the behaviour of the vehicle on the surface (1408). Controlling the vehicle over the surface (1410). The system may also receive data from a network outside of the vehicle and from other vehicles (V2V).

Driving scenario sampling for training/tuning machine learning models for vehicles

A method involving sampling driving scenarios for training machine learning models (603, fig 6) comprises: assigning, using at least one processor, a set of initial physical states to a set of objects in a map for a set of simulated driving scenarios, wherein the set of initial physical states are assigned according to one or more outputs of a random number generator; generating, using the at least one processor, the set of simulated driving scenarios in the map using the initial physical states of the objects in the set of objects; selecting, using the at least one processor, samples of the simulated driving scenarios; training, using the at least one processor, a machine learning model using the selected samples; and operating, using a control circuit, a vehicle in an environment using the trained machine learning model. The model may also use the drivers mental state in the simulation.

Road surface condition guided decision making and prediction

A road surface detection system comprising, receiving from at least one sensor of a vehicle, sensor data associated with a surface along a path to be travelled by a vehicle 1402. Using a surface classifier to determine a classification of the surface based on the sensor data 1404. Classifying the surface as an unknown surface and using sensor data to assign a classification to the unknown surface. Determining, based on the classification of the surface, drivability properties of the surface 1406. Planning, based on the drivability properties of the surface, a behaviour of the vehicle when driving near the surface or on the surface 1408 and controlling the vehicle based on the planned behaviour 1410. The system may generate a surface map and use V2V or V2I communications.