The invention relates to a traffic situation prediction method based on a cross attention neural network. According to the method, a space-time cross attention STCANN module is introduced, and a flow and speed space-time block and a cross attention mechanism are combined to capture a space-time relationship and features in traffic flow data. The STCANN module is composed of stacked space-time cross attention modules and a classification prediction layer, each stacked space-time cross attention module is composed of a space cross attention block and a time cross attention block, and the STCANN module can be used for carefully extracting space-time features in a road network and achieving deeper learning through the stacked modules. The classification prediction layer is composed of two convolution layers, multi-step prediction is achieved by combining the spatial-temporal features of the last cross attention module, and the final output is the classified road service level. The method shows uniqueness, innovativeness and superiority in a traffic situation prediction task.

    本发明涉及一种基于交叉注意力神经网络的交通态势预测方法。该方法通过引入时空交叉注意力STCANN模块,结合流量和速度时空块和交叉注意力机制,来捕捉交通流量数据中的时空关系和特征。STCANN模块由堆叠的时空交叉注意力模块和分类预测层组成,其中,每个堆叠的时空交叉注意模块由空间交叉注意力块和时间交叉注意块组成,可以用于细致地提取路网中的时空特征,并通过堆叠的模块来实现更深层次的学习。分类预测层由两个卷积层组成,结合最后一个交叉注意力模块的时空特征实现多步预测,并且最后的输出是分类的道路服务水平。该方法在交通态势预测任务中展现了独特性、创新性和优越性。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Traffic situation prediction method based on cross attention neural network


    Weitere Titelangaben:

    一种基于交叉注意力神经网络的交通态势预测方法


    Beteiligte:
    SHI BENYUN (Autor:in) / ZHANG XINRU (Autor:in) / GUO QI (Autor:in)

    Erscheinungsdatum :

    21.06.2024


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung



    Traffic prediction method based on attention reconstruction neural network

    CAI JIANGHUI / YANG HAIFENG / LIU AIQIN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    TRAFFIC SITUATION PREDICTION DEVICE, AND TRAFFIC SITUATION PREDICTION METHOD

    AOKI YASUHIRO / SUZUKI YOSHIHIKO / NARUSE KOSUKE | Europäisches Patentamt | 2023

    Freier Zugriff

    TRAFFIC SITUATION PREDICTION DEVICE AND TRAFFIC SITUATION PREDICTION METHOD

    NARUSE KOSUKE / UENO HIDEKI / OBA YOSHIKAZU | Europäisches Patentamt | 2024

    Freier Zugriff

    TRAFFIC SITUATION PREDICTION DEVICE, TRAFFIC SITUATION PREDICTION METHOD AND PROGRAM

    UEISHI ISAO / NAKAMURA KAZUKI / ABE NAOKI et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Traffic flow prediction method based on time-space diagram attention neural network

    LI BAILIN / WEN MI | Europäisches Patentamt | 2025

    Freier Zugriff