The invention discloses a distributed traffic flow prediction method and system based on space-time decoupling, and the method comprises the steps: carrying out the partitioning of an urban road network, deploying an edge cloud in each region, and uploading the collected traffic information to the edge cloud corresponding to the region through a sensor disposed on a road; the edge cloud and the core cloud jointly predict future traffic conditions through a neural network training model of space-time decoupling; wherein the edge cloud stores road traffic original data, the core cloud stores a road network topological relation, spatial and temporal feature extraction is decoupled, the edge cloud and the core cloud are deployed on the edge cloud and the core cloud respectively, the edge cloud obtains time features of traffic information, and the core cloud obtains spatial features of the traffic information; and edge clouds are utilized to collaboratively embed spatial information into the edge clouds to carry out fusion of spatial and temporal characteristics. Compared with a traditional method, under the edge cloud cooperation mechanism, the method reduces transmission delay and calculation pressure, fully extracts and fuses spatio-temporal features, and can improve the timeliness and accuracy of prediction.

    本发明公开了一种基于时空解耦的分布式交通流预测方法与系统,方法包括:对城市路网进行分区,每个区域部署一个边缘云,部署在道路上的传感器将收集到的交通信息上传到本区域对应的边缘云;边缘云联合核心云共同通过时空解耦的神经网络训练模型,预测未来的交通状况;其中边缘云存储道路交通原始数据,核心云存储路网拓扑关系,将时空特征提取进行解耦,分别部署在边缘云和核心云上,由边缘云获取交通信息的时间特征,核心云获取交通信息的空间特征,并利用边云协同在边缘云嵌入空间信息进行时空特征的融合。相比于传统方法,本发明在边云协同的机制下,降低了传输时延和计算压力,充分提取并融合了时空特征,能够提高预测的时效性和准确度。


    Access

    Download


    Export, share and cite



    Title :

    Distributed traffic flow prediction method and system based on space-time decoupling


    Additional title:

    一种基于时空解耦的分布式交通流预测方法与系统


    Contributors:
    ZHOU HAIBO (author) / ZHANG YUAN (author) / XU YUNTING (author) / LIU TINGTING (author)

    Publication date :

    2023-04-14


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    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 / H04L TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION , Übertragung digitaler Information, z.B. Telegrafieverkehr



    Space-time heterogeneous decoupling method based on multi-mode traffic flow prediction

    SENG DEWEN / WU MEI / LIN HUAI | European Patent Office | 2024

    Free access

    Traffic index prediction method based on decoupling space-time neural network

    HO CHIN KEONG / ZHANG JIPENG | European Patent Office | 2024

    Free access

    Traffic flow prediction method based on decoupling fusion network

    CHEN JING / PAN SHIXIANG | European Patent Office | 2024

    Free access

    Efficient space-time traffic flow prediction method

    WANG XITE / WU QIONG / BAI MEI et al. | European Patent Office | 2024

    Free access

    Online space-time traffic flow prediction method

    XIAO JIANHUA / XIAO HONGBO / DING LIMING et al. | European Patent Office | 2024

    Free access