The invention relates to a traffic prediction method based on dual dynamic space-time diagram convolution, which is used for solving the problem of low prediction precision caused by lack of opposite side modeling and dynamic modeling in a current traffic prediction method based on a graph network. Firstly, historical traffic data is input and sent to an input layer for processing, then output of the input layer is sent to a dynamic space-time layer, space-time correlation feature extraction is carried out through a plurality of stacked dual dynamic space-time blocks in the dynamic space-time layer, features are input to an output layer, and finally, a final prediction result is output. Wherein the core dual dynamic space-time block and the key dual dynamic space-time block comprise a dynamic graph convolution module, a dynamic hypergraph convolution module and a dynamic interaction module between the dynamic graph convolution module and the dynamic hypergraph convolution module. According to the method, the complex space-time correlation in the traffic data can be well mined, so that the potential space-time correlation of the dynamic traffic system is revealed, and the urban traffic data can be predicted more accurately.

    本发明涉及一种基于对偶动态时空图卷积的交通预测方法,用于解决当前基于图网络的交通预测方法中存在缺少对边建模以及动态建模导致的预测精度不高的问题。首先输入历史交通数据,送到输入层进行处理,然后将输入层的输出送入动态时空层,经过动态时空层中多个堆叠的对偶动态时空块进行时空相关性特征抽取,再将这些特征输入到输出层,最后输出的即是最终的预测结果。其中,最核心和关键的对偶动态时空块包括动态图卷积模块、动态超图卷积模块,以及两个之间的动态交互模块。本发明能很好的挖掘交通数据中复杂的时空相关性,从而揭示动态交通系统潜在的时空关联,进而更加准确的对城市交通数据进行预测。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Traffic prediction method based on dual dynamic space-time diagram convolution


    Weitere Titelangaben:

    一种基于对偶动态时空图卷积的交通预测方法


    Beteiligte:
    SUN YANFENG (Autor:in) / JIANG XIANGHENG (Autor:in) / HU YONGLI (Autor:in) / GUO KAN (Autor:in) / YIN BAOCAI (Autor:in)

    Erscheinungsdatum :

    2022-05-13


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / 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



    Traffic prediction method and device based on dynamic space-time diagram convolution attention model

    ZHANG TINGTING / LI RUI | Europäisches Patentamt | 2021

    Freier Zugriff

    Attention-based dynamic space-time diagram convolution traffic flow prediction system and method

    JI ZHENYUAN / HUAN YUEHUI / YANG GENKE et al. | Europäisches Patentamt | 2023

    Freier Zugriff


    Traffic flow prediction method and system based on trend space-time diagram convolution, and medium

    ZONG XINLU / YU FAN / WANG CHUNZHI et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Traffic flow prediction method of recurrent neural network based on dynamic diffusion diagram convolution

    LI YUN / TIAN YUDOU / JING PEIGUANG | Europäisches Patentamt | 2022

    Freier Zugriff