The invention discloses a traffic signal lamp control system and method based on reinforcement learning and dynamic timing, and the method comprises the steps: carrying out the modeling of a crossroad, and respectively defining the state, the reward and the action of reinforcement learning; constructing a strategy network and an evaluation network according to a reinforcement learning A2C algorithm; and finally, collecting intersection real-time information, converting the intersection real-time information into state vectors, inputting the state vectors into the constructed strategy network and evaluation network, outputting action probability distribution, and selecting the most suitable phase by the system according to the action probability distribution; wherein after the phase is determined, the next period of time belongs to the green light time of the phase; enabling a dynamic timing module formed by a self-adaptive timing controller dynamically to control the green light time according to real-time traffic flow information by taking maximum green light time utilization as a target; when the loss of the green light time is detected, terminating the current phase. According to the invention, the defect of low green light utilization rate caused by fixed traffic light time distribution is overcome, and the average waiting time of vehicles and the emission of harmful gases such as carbon dioxide are reduced.

    本发明公开了一种基于强化学习与动态配时的交通信号灯控制系统及方法,首先对十字路口进行建模,分别定义强化学习三要素state,reward,action;然后依据强化学习A2C算法构建策略网络和评估网络;最后收集路口实时信息,转化为state向量后输入到构建的策略网络和评估网络,输出action概率分布,系统根据action概率分布选择最合适的相位;决定相位后,接下来的一段时间就属于该相位的绿灯时间;由自适应定时控制器构成的动态配时模块根据实时的交通流信息,以最大化绿灯时间利用为目标,动态控制绿灯时间;当检测到绿灯时间的损失,就终止当前相位。本发明克服了固定分配交通灯时长而导致绿灯利用率不高的缺陷,减少了车辆的平均等待时间和二氧化碳等有害气体的排放。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Traffic signal lamp control system and method based on reinforcement learning and dynamic timing


    Weitere Titelangaben:

    基于强化学习与动态配时的交通信号灯控制系统及方法


    Beteiligte:
    LIU BINGYI (Autor:in) / SONG BIN (Autor:in) / WU PENGCHUN (Autor:in) / XIONG SHENGWU (Autor:in)

    Erscheinungsdatum :

    2021-12-07


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

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



    Traffic signal lamp timing method, device and equipment based on deep reinforcement learning

    ZHOU JIANHONG / HUANG YUMIN / WANG YUNXIANG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Traffic signal lamp control method based on migration reinforcement learning

    ZHANG CHENGWEI / LIU XIN / LI YIHONG | Europäisches Patentamt | 2023

    Freier Zugriff

    Crossroad traffic signal lamp control method based on reinforcement learning

    WU LIBING / WANG MIN / ZHANG RUI et al. | Europäisches Patentamt | 2020

    Freier Zugriff

    Traffic signal lamp control method and system based on reinforcement learning

    LUO JUAN / ZHENG YANLIU | Europäisches Patentamt | 2021

    Freier Zugriff

    Traffic signal lamp timing method

    LI YING | Europäisches Patentamt | 2020

    Freier Zugriff