The invention discloses a vehicle multi-modal trajectory prediction method, system and device and a storage medium, and the method comprises the following steps: obtaining vehicle trajectory and lane data, extracting trajectory features and lane features through employing an attention model, constructing a multi-scale hypergraph, processing the trajectory features to obtain interaction features, and carrying out the prediction of the vehicle multi-modal trajectory. Transmitting the interaction features to the lane features, constructing a global probability network graph of the future driving route of the vehicle, and predicting the vehicle trajectory through the global probability network graph. The road test sensing unit is used for acquiring high-precision vehicle track data and lane data, so that feature extraction and calculation are facilitated; modeling is carried out on the interaction between the vehicles through the multi-scale hypergraph, and quantitative calculation and interpretable expression of the interaction between the vehicles are achieved. A global probability network diagram is introduced to represent the uncertainty of vehicle driving, and a plurality of predicted vehicle tracks are obtained by combining an algorithm and a track decoder, so that accurate and multi-modal vehicle track prediction is realized.

    本发明公开了一种车辆多模态轨迹预测方法、系统、设备及存储介质,所述车辆多模态轨迹预测方法包括以下步骤:获取车辆轨迹和车道数据,使用注意力模型提取轨迹特征和车道特征,构建多尺度超图处理轨迹特征得到交互特征,将交互特征传递给车道特征并构建车辆未来行驶路线的全局概率网络图,通过全局概率网络图对车辆轨迹进行预测。使用路测感知单元获取高精度的车辆轨迹数据和车道数据,便于特征的提取与计算;通过多尺度超图对车辆之间的相互作用进行建模,实现了车辆之间相互作用的量化计算和可解释性表达;引入全局概率网络图表征车辆驾驶的不确定性,结合采用算法、轨迹解码器得到预测的多条车辆轨迹,实现准确、多模态的车辆轨迹预测。


    Access

    Download


    Export, share and cite



    Title :

    Vehicle multi-modal trajectory prediction method, system and device and storage medium


    Additional title:

    一种车辆多模态轨迹预测方法、系统、设备及存储介质


    Contributors:
    ZHAO CONG (author) / ZOU YING (author) / LING QIANG (author) / ZENG ZIMU (author) / JI YUXIONG (author) / DU YUCHUAN (author) / XU DONG (author) / CAO PENGLIANG (author) / WANG WEI (author) / WANG XIAOLI (author)

    Publication date :

    2024-09-10


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Multi-modal trajectory prediction method

    JIANG WENJUAN / JIN ZHI / WANG REN et al. | European Patent Office | 2023

    Free access

    Multi-modal trajectory prediction method based on target driving, vehicle, equipment and medium

    LU LIPING / YE PEICHANG | European Patent Office | 2024

    Free access

    Vehicle trajectory prediction method, system and device and storage medium

    YIN YACHENG / HOU WENLEI / GUAN JUNYU et al. | European Patent Office | 2024

    Free access

    Multi-modal multi-agent trajectory prediction

    SUN PEI / ZHAO HANG / MCCAULEY ALEXANDER et al. | European Patent Office | 2024

    Free access

    MULTI-MODAL MULTI-AGENT TRAJECTORY PREDICTION

    SUN PEI / ZHAO HANG / MCCAULEY ALEXANDER et al. | European Patent Office | 2022

    Free access