Techniques are described for estimating intent of pedestrians and other road users in the vicinity of a vehicle. In some embodiments, the techniques include obtaining, by a computer system of a vehicle equipped with one or more sensors, a sequence of video frames corresponding to a scene external to the vehicle, detecting one or more vulnerable road users (VRUs) in the sequence of video frames, where the detecting includes estimating a gesture of each of the detected one or more VRUs. The technique further includes generating a segmented map of the scene using the one or more video frames; one or more intent probabilities are estimated using the estimated poses of the one or more VRUs and the segmentation map, each intent probability corresponding to one of the detected one or more VRUs, and one or more automated driving action probabilities are adjusted based on the estimated one or more intent.
描述了用于估计车辆附近的行人和其他道路使用者的意图的技术。在某些实施例中,该技术包括通过配备一个或多个传感器的车辆的计算机系统获得对应于车辆外部场景的视频帧序列,检测视频帧序列中的一个或多个易受伤害的道路使用者(VRUs),其中检测包括估计检测到的一个或多个VRUs中的每一个的姿势。该技术还包括使用一个或多个视频帧生成场景的分割图;使用一个或多个VRUs的估计姿势和分割图估计一个或多个意图概率,每个意图概率对应于检测到的一个或多个VRUs之一,并基于估计的一个或多个意图调节一个或多个自动驾驶动作概率。
Method and apparatus for predicting intent of vulnerable road user
用于预测易受伤害道路使用者的意图的方法和装置
2022-09-09
Patent
Electronic Resource
Chinese
METHOD AND APPARATUS FOR PREDICTING INTENT OF VULNERABLE ROAD USERS
European Patent Office | 2021
|Method and apparatus for predicting intent of vulnerable road users
European Patent Office | 2023
|METHOD AND APPARATUS FOR PREDICTING INTENT OF VULNERABLE ROAD USERS
European Patent Office | 2022
|METHOD AND APPARATUS FOR PREDICTING INTENT OF VULNERABLE ROAD USERS
European Patent Office | 2021
|