The invention discloses an urban intersection signal control method based on deep reinforcement learning. The method comprises the following steps: acquiring traffic flow information in a road network environment; constructing a value network and a target network, defining a state space by adopting a discretization modeling method of a traffic environment state, defining an action space by adopting a self-adaptive signal timing method of an unfixed phase sequence, and determining a reward function according to the change of vehicle waiting time before and after action execution; performing action selection according to the traffic flow information in the road network environment by using the value network, and calculating the action with the maximum value corresponding to the next state; calculating the action value of the action according to the action with the maximum value by utilizing the target network; and performing network parameter updating on the value network and the target network by adopting a gradient descent method to obtain a final urban intersection signal control result. The method can effectively improve the signal control accuracy and processing efficiency of the urban intersection, thereby improving the passing efficiency of the intersection, and reducing the average delay and average queuing length of vehicles at the intersection.

    本发明公开了一种基于深度强化学习的城市交叉口信号控制方法,包括获取路网环境中的车流信息;构建价值网络和目标网络,采用交通环境状态的离散化建模方法定义状态空间,采用不固定相序的自适应式信号配时方法定义动作空间,根据动作执行前后车辆等待时间的变化确定奖励函数;利用价值网络根据路网环境中的车流信息进行动作选择,计算下一状态对应的价值最大的动作;利用目标网络根据价值最大的动作计算该动作的动作价值;采用梯度下降方法对价值网络和目标网络进行网络参数更新,得到最终的城市交叉口信号控制结果。本发明能够有效提高城市交叉口信号控制准确性和处理效率,从而提高交叉口通行效率,降低交叉口车辆的平均延误和平均排队长度。


    Access

    Download


    Export, share and cite



    Title :

    Urban intersection signal control method based on deep reinforcement learning


    Additional title:

    基于深度强化学习的城市交叉口信号控制方法


    Contributors:
    JU YANNI (author) / MA DIE (author)

    Publication date :

    2024-06-11


    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



    Multi-intersection signal cooperative control method based on deep reinforcement learning

    MA CHANGXI / LIU YAN / ZHAO HONGXING et al. | European Patent Office | 2025

    Free access

    Multi-intersection traffic signal control method based on deep reinforcement learning

    DENG HENG / WANG YULONG / GAO YANG et al. | European Patent Office | 2024

    Free access

    Multi-intersection traffic signal control method based on deep reinforcement learning

    LIU LIJUAN / BAI GUANGMING | European Patent Office | 2023

    Free access

    Single-intersection signal control method based on deep reinforcement learning algorithm

    HUANG YIWANG / WU QIAN | European Patent Office | 2023

    Free access