The invention provides a traffic prediction method and system based on self-supervised learning. The method comprises the following steps: acquiring traffic flow data; preprocessing the traffic flow data; constructing a traffic prediction model based on the neural network model; capturing space-time dependence of the traffic flow data to obtain original space-time embedding; performing data masking processing on the traffic flow training data to obtain enhanced traffic flow data; capturing the space-time dependence of the enhanced traffic flow data to obtain enhanced space-time embedding; then respectively obtaining a spatial heterogeneity loss function and a time heterogeneity loss function; traffic flow prediction is carried out through a multi-layer perceptron, and a prediction loss function is obtained; finally calculating to obtain a total loss function of the model; optimizing model parameters of the traffic prediction model according to the model total loss function; and inputting the traffic flow test data into the traffic prediction model to obtain a traffic prediction result of the target city. The method has the effect of accurately predicting the urban traffic condition.

    本发明提供一种基于自监督学习的交通预测方法及系统,该方法包括如下步骤:获取交通流量数据;预处理交通流量数据;基于神经网络模型构建交通预测模型;捕捉交通流量数据的时空依赖性,得到原始时空嵌入;对交通流量训练数据进行数据掩盖处理,得到增强交通流量数据;捕捉增强交通流量数据的时空依赖性,得到增强时空嵌入;然后分别得到空间异质性损失函数和时间异质性损失函数;通过多层感知机进行交通流量预测,获取预测损失函数;最终计算得到模型总损失函数;根据模型总损失函数优化交通预测模型的模型参数;将交通流量测试数据输入至交通预测模型中,得到目标城市的交通预测结果。本发明具有较为准确预测城市交通情况的效果。


    Access

    Download


    Export, share and cite



    Title :

    Traffic prediction method and system based on self-supervised learning


    Additional title:

    基于自监督学习的交通预测方法及系统


    Contributors:
    WEI YINGMEI (author) / GAO MIN (author) / CHEN ZIMO (author) / ZHANG YITONG (author) / XIE YUXIANG (author) / GUO YANMING (author) / JIANG JIE (author) / KANG LAI (author)

    Publication date :

    2024-07-02


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    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 / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Self-Supervised Traffic Advisors: Distributed, Multi-view Traffic Prediction for Smart Cities

    Sun, Jiankai / Kousik, Shreyas / Fridovich-Keil, David et al. | IEEE | 2022



    Short-Term Traffic Flow Prediction on Highways Based on Self-Supervised Spatio-Temporal Transformer

    Guo, Xingping / Song, Jingni / Du, Kai et al. | Springer Verlag | 2025


    Supervised Weighting-Online Learning Algorithm for Short-Term Traffic Flow Prediction

    Jeong, Young-Seon / Byon, Young-Ji / Castro-Neto, Manoel Mendonca et al. | IEEE | 2013


    Traffic flow prediction method based on variational self-coding learning

    FENG XUAN / CAI YOUBAO / JIAO WEICHAO et al. | European Patent Office | 2022

    Free access