The invention discloses a traffic flow prediction method based on a generative adversarial network. The method comprises the steps: firstly constructing a traffic physical topological structure graphnetwork according to the spatial topological structure of a road segment, and then constructing an initial neighborhood traffic graph network and a high-order neighborhood traffic graph network according to the connection relation of nodes in a traffic graph; and then performing feature extraction on nodes in the domain of the nodes by using a graph attention network (GAT) constructed by the method. According to the prediction process, a predicted traffic flow value is generated by using a generator of a generative adversarial network (GAN) constructed by the method, and adversarial training is performed on the predicted traffic flow value and a real traffic flow value by using a discriminator of the generative adversarial network (GAN) constructed by the method, so that the predicted traffic flow value and the real traffic flow value play each other and advance each other. According to the steps, the parameter optimization model is adjusted step by step, and finally an optimal trafficflow model is obtained. According to the invention, the traffic flow value closest to the real value is generated by using the model, so that the purpose of traffic flow prediction is achieved.

    一种基于生成式对抗网络的交通流预测方法,先根据路段的空间拓扑结构构建交通物理拓扑结构图网络,接着根据交通图中节点的连接关系构建初始邻域交通图网络与高阶邻域交通图网络,接着用本发明构建好的图注意力网络(GAT)对节点的领域内节点进行特征提取。预测过程为利用本发明构建的生成式对抗网络(GAN)的生成器生成预测的交通流量值,并利用本发明构建的生成式网络(GAN)的判别器对预测的交通流量值与真实的交通流量值进行对抗训练,使得两者相互博弈,相互进步。根据以上步骤,并逐步调整参数优化模型,最终得到一个最佳的交通流量模型。本发明利用此模型生成最接近真实值的交通流量值,以达到交通流量预测的目的。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Traffic flow prediction method based on generative adversarial network


    Weitere Titelangaben:

    一种基于生成式对抗网络的交通流预测方法


    Beteiligte:
    XU DONGWEI (Autor:in) / LIN ZHENQIAN (Autor:in) / YANG HAO (Autor:in) / GU CHUNTAO (Autor:in) / DING JIALI (Autor:in) / ZHOU LEI (Autor:in) / XUAN QI (Autor:in)

    Erscheinungsdatum :

    2021-02-05


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

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



    Urban traffic flow prediction method based on generative adversarial network

    ZHU NINGBO / PU BIN / FAN XINXIN et al. | Europäisches Patentamt | 2020

    Freier Zugriff

    SATP-GAN: self-attention based generative adversarial network for traffic flow prediction

    Zhang, Liang / Wu, Jianqing / Shen, Jun et al. | Taylor & Francis Verlag | 2021


    Urban traffic flow prediction method, system and equipment based on generative adversarial network

    WEI ZHICHENG / ZHANG TAOYI | Europäisches Patentamt | 2023

    Freier Zugriff

    Road traffic network state prediction based on a generative adversarial network

    Xu, Dongwei / Peng, Peng / Wei, Chenchen et al. | Wiley | 2020

    Freier Zugriff

    Road traffic network state prediction based on a generative adversarial network

    Xu, Dongwei / Peng, Peng / Wei, Chenchen et al. | IET | 2020

    Freier Zugriff