The invention relates to a method (100) for predicting a movement of at least one traffic-related object (1) on the basis of observed values of an environment (2) of the at least one traffic-related object (1), comprising the following steps: mapping (110) observed values O [tau] 0 of the environment (2) at a point in time [tau] 0 by a trained encoder network (3) onto a dimensionality-reduced representation z [tau] 0; determining (120), using at least one trained prediction network (4, 5), from the action atau0 performed by the object (1) at the point in time tau0 and the representation ztau0:-for a prediction # imgabs0 # representing ztau1, where the trained encoder network (3) is expected to map future observations otau1 at the point in time tau1 > tau0 to the representation,-for a prediction # imgabs0 # representing ztau0, and-for a prediction # imgabs0 # imgabs0 # imgabs0 # imgabs0 # imgabs0 # imgabs0 # imgabs0 # imgabs0; and/or-a prediction # imgabs1 # of an action atau 1 expected to be performed by the subject (1) at a point in time tau 1. A prediction # imgabs4 # of a dynamic state d tau 1 of the subject (1) at a point in time tau 1 is determined based on the prediction # imgabs2 # and/or # imgabs3 # (see formula (1)) (130).

    本发明涉及一种用于基于至少一个交通相关对象(1)的环境(2)的观察值来预测所述至少一个交通相关对象(1)的运动的方法(100),具有以下步骤:·由经过训练的编码器网络(3)将在时间点τ0的所述环境(2)的观察值Oτ0映射到降维的表示zτ0上(110);·根据由所述对象(1)在时间点τ0实施完成的动作aτ0和表示zτ0,使用至少一个经过训练的预测网络(4、5)而确定(120):‑对表示zτ1的预测#imgabs0#其中经过训练的所述编码器网络(3)预计将在时间点τ1>τ0的未来观察值oτ1映射到所述表示,和/或‑对所述对象(1)在时间点τ1预计执行的动作aτ1的预测#imgabs1#·基于所述预测#imgabs2#和/或#imgabs3#确定对所述对象(1)在时间点τ1的动态状态的dτ1的预测#imgabs4#(参见公式(1))(130)。


    Access

    Download


    Export, share and cite



    Title :

    Motion prediction of traffic participants


    Additional title:

    交通参与者的运动预测


    Contributors:

    Publication date :

    2024-04-05


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06V



    PROBABILISTIC MAP-BASED PEDESTRIAN MOTION PREDICTION TAKING TRAFFIC PARTICIPANTS INTO CONSIDERATION

    Wu, Jingyuan / Ruenz, Johannes / Althoff, Matthias | British Library Conference Proceedings | 2018



    Road occupancy prediction of traffic participants

    Althoff, Matthias / Hess, Daniel / Gambert, Florian | IEEE | 2013