Trajectory prediction using deep learning multiple predictor fusion and bayesian optimization
Номер патента: EP3924901A1
Опубликовано: 22-12-2021
Автор(ы): Gaowei Xu, Maximilian DOEMLING, Niklas HAUSER, Yao Ge
Принадлежит: Bayerische Motoren Werke AG
Опубликовано: 22-12-2021
Автор(ы): Gaowei Xu, Maximilian DOEMLING, Niklas HAUSER, Yao Ge
Принадлежит: Bayerische Motoren Werke AG
Реферат: A method, an apparatus, and a vehicle for trajectory prediction are provided. The method comprises obtaining observation data for an object and environment information related to the object; giving respective predictions for the object based on historical observation data with a plurality of predictors; feeding the respective predictions and the environment information into a trained multiple predictor fusion model for trajectory prediction; providing, by the multiple predictor fusion model, a predicted trajectory for the object under the environment indicated by the environment information.
Method and system for delay prediction for scheduled public transport using multi architectural deep learning
Номер патента: US20240112096A1. Автор: Rajesh Jayaprakash,Arvind RAMANUJAM,Rohith Regikumar,Priyanga Kasthurirajan. Владелец: Tata Consultancy Services Ltd. Дата публикации: 2024-04-04.