The invention belongs to the technical field of intelligent traffic, and particularly relates to a traffic state prediction method based on an adaptive dynamic space-time diagram convolutional network, which comprises the following steps: firstly, adaptively deducing a diagram structure from macroscopic and microscopic angles; secondly, extracting a time dependency relationship by using a multi-scale gating TCN; thirdly, sensing a historical change trend hidden in the traffic state data sequence through a time trend sensing self-attention mechanism, thereby realizing accurate long-term prediction; and finally, a dynamic filter is generated at each time step to filter node embedding, a dynamic graph is generated, and dynamic spatial dependency is captured by combining an adaptive adjacency matrix. According to the method, the complex space-time correlation in the traffic state data can be well mined, so that the potential space-time correlation of the dynamic traffic system is revealed. Wide experiments are carried out in two real traffic state data sets, and experiment results show that the method achieves a good prediction level.

    本发明属于智能交通技术领域,具体涉及一种基于自适应动态时空图卷积网络的交通状态预测方法,该方法包括:首先,从宏观和微观两个角度自适应地推断图结构;其次,使用多尺度门控TCN提取时间依赖关系;再次,通过时间趋势感知自注意机制感知隐藏在交通状态数据序列中的历史变化趋势,从而实现准确的长期预测;最后,在每个时间步生成动态过滤器对节点嵌入进行过滤,生成动态图,通过结合自适应邻接矩阵捕获动态的空间依赖性。本发明能很好的挖掘交通状态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。在两个真实的交通状态数据集中进行了广泛的实验,实验结果表明本发明方法达到了很好的预测水平。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Traffic state prediction method based on adaptive dynamic space-time diagram convolutional network


    Weitere Titelangaben:

    一种基于自适应动态时空图卷积网络的交通状态预测方法


    Beteiligte:
    ZHUANG XUFEI (Autor:in) / MAO RUI (Autor:in) / GAO XUDONG (Autor:in) / ZHANG HAITAO (Autor:in) / WANG YUJIE (Autor:in) / LI ZIHENG (Autor:in) / DU TING (Autor:in) / ZHAO CHANCHAN (Autor:in)

    Erscheinungsdatum :

    31.05.2024


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

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



    Traffic flow prediction method based on interactive adaptive space-time diagram convolutional network

    SHI QUAN / CAO CHENYANG / BAO YINXIN et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow prediction method based on space-time diagram convolutional network

    JIANG YUNLIANG / XIA RENHUAN / ZHANG XIONGTAO et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Traffic flow prediction method based on space-time diagram convolutional network

    JIANG CONG / SONG YUN / DENG ZELIN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Traffic flow prediction method of space-time diagram convolutional network

    TENG FEI / WANG ZIDAN / QIAO LU et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow prediction method based on multi-space-time diagram convolutional network

    SHI QUAN / DAI JUNMING / SHEN QINQIN et al. | Europäisches Patentamt | 2021

    Freier Zugriff