The invention provides a vehicle trajectory prediction method based on dynamic interaction graph convolution, which comprises the following steps of: firstly, preprocessing a vehicle trajectory, then performing composition according to vehicle distribution of different time frames in a traffic road scene, capturing potential interaction information existing between a target vehicle and surrounding vehicles through a dynamic graph convolution network module, and predicting the vehicle trajectory according to the potential interaction information. Wherein an interaction weight matrix between different vehicles is dynamically updated by adopting a long-short-term memory network, historical track information of a target vehicle is learned through a Transform network, and space interaction characteristics and time sequence track characteristics are combined to serve as encoder input to generate vehicle track coordinates at the next moment. According to the method, a new dynamic graph convolution method is provided, the vehicles serve as nodes, the weight information between the vehicles is updated in real time, the defect that interaction information between the vehicles and surrounding objects cannot be dynamically obtained in a traditional vehicle track prediction method is overcome, and the prediction precision of the vehicle track is greatly improved.

    本发明提供了一种基于动态交互图卷积的车辆轨迹预测方法,首先将车辆轨迹进行预处理,然后根据交通道路场景中不同时间帧的车辆分布进行构图,通过动态图卷积网络模块捕捉目标车辆与周围车辆之间存在的潜在交互信息,其中不同车辆之间的交互权重矩阵采用长短期记忆网络动态更新,目标车辆的历史轨迹信息通过Transformer网络进行学习,通过结合空间交互特征以及时序轨迹特征作为编码器输入,以生成下一时刻的车辆轨迹坐标。本发明通过提出一个新的动态图卷积方法,将车辆作为节点并实时更新车辆间的权重信息,解决了传统车辆轨迹预测方法中无法动态获取与周围对象之间交互信息的缺点,大大提高了车辆轨迹的预测精度。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Vehicle track prediction method based on dynamic interaction graph convolution


    Weitere Titelangaben:

    一种基于动态交互图卷积的车辆轨迹预测方法


    Beteiligte:
    SHEN GUOJIANG (Autor:in) / LI PENGFEI (Autor:in) / KONG XIANGJIE (Autor:in) / ZHENG JIANWEI (Autor:in) / LIU ZHI (Autor:in)

    Erscheinungsdatum :

    29.07.2022


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / 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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung



    Traffic prediction method based on dynamic graph convolution

    FAN JIN / WENG WENCHAO / TIAN HAO et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Dynamic graph convolution traffic speed prediction method

    LIU QILIANG / YUAN HAOTAO / YANG LIU et al. | Europäisches Patentamt | 2020

    Freier Zugriff

    Traffic flow prediction method based on dynamic sparse graph convolution GRU

    ZHANG LINLIANG / YIN JIALI / LI SHUO et al. | Europäisches Patentamt | 2025

    Freier Zugriff

    Traffic Flow Prediction Based on Dynamic Time Slot Graph Convolution

    Chen, Hongwei / Wang, Han / Chen, Zexi | Transportation Research Record | 2025


    Dual Dynamic Spatial-Temporal Graph Convolution Network for Traffic Prediction

    Sun, Yanfeng / Jiang, Xiangheng / Hu, Yongli et al. | IEEE | 2022