The invention provides an urban multi-type traffic participant multi-modal vehicle trajectory prediction method based on a graph traversal strategy. The method comprises the steps of obtaining information data of a target vehicle, surrounding multiple traffic participants and lane nodes in a preset time period; inputting the information data into a preset trajectory prediction model to obtain a final prediction trajectory of the target vehicle: performing feature extraction and aggregation on the information data to obtain lane node aggregation features; performing graph traversal strategy processing on the lane node aggregation features to obtain future node trajectory distribution of the vehicle; and performing multi-modal prediction on the future node trajectory distribution of the vehicle to obtain a final prediction trajectory of the target vehicle. According to the method, multiple types of traffic participants in a complex urban traffic scene are considered, a trajectory prediction model based on a graph traversal strategy is provided based on historical trajectory information, and compared with a traditional trajectory prediction model, the trajectory prediction model is closer to the real situation, and the trajectory prediction precision is effectively improved.

    本发明提出了基于图遍历策略的城市多类型交通参与者多模态车辆轨迹预测方法,包括:获取预设时间段内目标车辆、周围多交通参与者和车道节点各自的信息数据;将所述信息数据输入预设的轨迹预测模型,获取目标车辆的最终预测轨迹:对所述信息数据进行特征提取及聚合,获取车道节点聚合特征;对所述车道节点聚合特征进行图遍历策略处理,获取车辆未来的节点轨迹分布;对所述车辆未来的节点轨迹分布,进行多模态预测,获取目标车辆的最终预测轨迹。本发明考虑复杂城市交通场景下多类型交通参与者,基于历史轨迹信息,提出基于图遍历策略的轨迹预测模型,相比传统轨迹预测模型更加接近真实情况,并有效提升预测轨迹的精度。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Multi-modal vehicle trajectory prediction method for urban multi-type traffic participants based on graph traversal strategy


    Weitere Titelangaben:

    基于图遍历策略的城市多类型交通参与者多模态车辆轨迹预测方法


    Beteiligte:
    ZHOU YIWEI (Autor:in) / LAO YINAN (Autor:in) / CHAMOT (Autor:in)

    Erscheinungsdatum :

    22.11.2024


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / 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 / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Multi-modal vehicle trajectory prediction method based on graph interaction mechanism

    GAO JIANPING / YANG YU / LIU PAN et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Multi-modal trajectory prediction method

    JIANG WENJUAN / JIN ZHI / WANG REN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    MULTI-MODAL MULTI-AGENT TRAJECTORY PREDICTION

    SUN PEI / ZHAO HANG / MCCAULEY ALEXANDER et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Multi-modal vehicle trajectory prediction based on mutual information

    Fei, Cong / He, Xiangkun / Ji, Xuewu | IET | 2020

    Freier Zugriff

    Multi-modal multi-agent trajectory prediction

    SUN PEI / ZHAO HANG / MCCAULEY ALEXANDER et al. | Europäisches Patentamt | 2024

    Freier Zugriff