The invention belongs to the technical field of intelligent traffic, and particularly relates to a signal lamp control system and method based on reinforcement learning and a self-attention mechanism. The system comprises a traffic data collection module, a traffic hidden state coding module, a traffic hidden state prediction module and a signal lamp phase control module. The data collection module records signal lamp phases and the total number of vehicles in each lane in a crossing road, and the hidden state encoding module encodes the traffic hidden state of the current crossing by using the current vehicle data and the current signal lamp phase of each lane. The hidden state prediction module predicts the traffic hidden state of the current intersection by using the historical time sequence data of the vehicles on each lane of the intersection and the historical time sequence data of the signal lamp phase; and the signal lamp phase control module decides the signal lamp phase according to the traffic hidden state of the intersection and displays a corresponding control signal. The traffic efficiency of the intersection is effectively improved by combining reinforcement learning; and the robustness of distorted traffic data is improved by using a self-attention mechanism.

    本发明属于智能交通技术领域,具体为基于强化学习和自注意力机制的信号灯控制系统及方法。本发明系统包括交通数据收集模块、交通隐状态编码模块、交通隐状态预测模块、信号灯相位控制模块;数据收集模块记录信号灯相位和交叉道路中各车道车辆总数,隐状态编码模块使用各车道当前车辆数据和当前信号灯相位对当前交叉路口的交通隐状态进行编码,隐状态预测模块使用交叉道路各车道车辆的历史时序数据和信号灯相位的历史时序数据对当前交叉路口的交通隐状态进行预测,信号灯相位控制模块根据交叉路口的交通隐状态决策信号灯相位并显示对应的控制信号。本发明结合强化学习有效提升交叉路口的通行效率;使用自注意力机制提升对失真交通数据的鲁棒性。


    Access

    Download


    Export, share and cite



    Title :

    Signal lamp control system and method based on reinforcement learning and self-attention mechanism


    Additional title:

    基于强化学习和自注意力机制的信号灯控制系统及方法


    Contributors:
    SUN WEIWEI (author) / QIN MINHAO (author)

    Publication date :

    2024-07-30


    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 / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Adaptive traffic signal control method based on reinforcement learning and self-attention mechanism

    ZHAO RUI / HU HAOFENG / FAN YUZE et al. | European Patent Office | 2024

    Free access

    Deep reinforcement learning traffic signal control method based on attention mechanism

    WU JIANGUANG / ZHOU SHUYA / HOU XIANGDONG et al. | European Patent Office | 2024

    Free access

    Traffic signal lamp control method and system based on reinforcement learning

    LUO JUAN / ZHENG YANLIU | European Patent Office | 2021

    Free access

    Traffic signal lamp control method based on migration reinforcement learning

    ZHANG CHENGWEI / LIU XIN / LI YIHONG | European Patent Office | 2023

    Free access

    Signal lamp intelligent control method based on deep reinforcement learning

    WEI KAI / ZHU YONG / ZHANG DONGHAI et al. | European Patent Office | 2024

    Free access