The invention discloses a vehicle trajectory prediction method and system based on a motion uncertainty diffusion model. The method comprises the steps of obtaining a real trajectory of a training vehicle; performing spatial-temporal feature coding processing on a past real track in the real tracks to obtain real track features; gaussian noise fusion processing is carried out based on the motion uncertainty diffusion principle, the real track and the real track characteristics, and a Gaussian noise removal algorithm is obtained through training; constructing a motion uncertainty diffusion model according to a Gaussian noise removal algorithm; when track prediction needs to be carried out on the to-be-predicted vehicle, obtaining known track characteristics of the to-be-predicted vehicle; and processing the known trajectory features based on the motion uncertainty diffusion model to obtain a predicted trajectory. And the vehicle trajectory prediction result can give consideration to diversity and accuracy.

    本发明公开了基于运动不确定性扩散模型的车辆轨迹预测方法及系统,方法包括:获取训练车辆的真实轨迹;对真实轨迹中的过去真实轨迹进行时空特征编码处理,得到真实轨迹特征;基于运动不确定性扩散原理、真实轨迹及真实轨迹特征进行高斯噪声融合处理,训练得到去高斯噪声算法;根据去高斯噪声算法构建得到运动不确定性扩散模型;当需要对待预测车辆进行轨迹预测时,获取待预测车辆的已知轨迹特征;基于运动不确定性扩散模型对已知轨迹特征进行处理,得到预测轨迹。使得车辆轨迹预测的结果能够兼顾多样性和准确性。


    Access

    Download


    Export, share and cite



    Title :

    Vehicle trajectory prediction method and system based on motion uncertainty diffusion model


    Additional title:

    基于运动不确定性扩散模型的车辆轨迹预测方法及系统


    Contributors:
    XIAO ZHU (author) / WU LINSHAN (author) / JIANG HONGBO (author) / YANG KEHUA (author) / LIU DAIBO (author)

    Publication date :

    2023-07-07


    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung



    Vehicle trajectory prediction method, trajectory prediction model training method and device

    SUN XINJIE / TIAN LEI / LIU YANG et al. | European Patent Office | 2023

    Free access

    Motion Query-based Multimodal Vehicle Trajectory Prediction for Autonomous Driving

    Jiang, Hao / Niu, Yixun / Hu, Chuan et al. | IEEE | 2023


    INTP-DM: Intention-Aware Multimodal Vehicle Trajectory Prediction Using Diffusion Model

    Yang, Yuanbo / Jin, Sheng / Na, Xiaoxiang et al. | IEEE | 2024


    An Enabling Decision-Making Scheme by Considering Trajectory Prediction and Motion Uncertainty

    Wang, Mingqiang / Zhang, Lei / Zhang, Zhiqiang et al. | IEEE | 2024


    SYSTEMS AND METHODS FOR UNCERTAINTY ESTIMATION IN VEHICLE TRAJECTORY PREDICTION

    MA HENGBO / HUYNH MANH TRUNG / CHOI CHIHO | European Patent Office | 2023

    Free access