The invention discloses a traffic light control method based on fog calculation and reinforcement learning. The traffic light control method comprises the steps of: firstly, acquiring intersection traffic conditions by means of a fog node of a present intersection, and broadcasting the intersection traffic conditions to fog nodes of adjacent intersections; acquiring a number of vehicles at the intersection and calculating green light time required for the vehicles to pass through the intersection; calculating green light time required for a vehicle stream of adjacent intersections to pass through the intersections; calculating green light time required for a traffic stream of the adjacent intersections to pass through the intersections; synthesizing numbers of vehicles at the present intersection and the adjacent intersections by means of the intersection and calculating total green light time required by the present intersection; and acting the calculated green light duration on a traffic signal lamp by means of an Agent of the present intersection, observing fed back reward signals, and transferring an environment state into a state at a next moment. The traffic light control method can allocate the green light time according to the real-time number of vehicles at the intersection, can reduce the parking time and the parking frequency, improves the road adjusting efficiency,and can effectively relieve the traffic jam.

    本发明公开了一种基于雾计算和强化学习的交通灯控制方法,首先本路口雾节点采集路口交通状况并广播给相邻路口的雾节点,采集路口车辆数目并计算车辆通过本路口所需绿灯时间,计算相邻路口的车流通过路口所需要的绿灯时间、相邻路口的交通流通过交叉路口所需要的绿灯时间、路口综合自身路口和相邻路口的车辆数目以及计算自身需要的总绿灯时间,本路口的Agent把计算的绿灯时长作用到交通信号灯上,观测反馈的奖赏信号,环境状态转移到下一时刻的状态,本发明能够根据十字路口的实时车辆数量分配绿灯时间,能减少停车时间和停车次数,提高道路调节效率,可以有效缓解交通拥堵。


    Access

    Download


    Export, share and cite



    Title :

    Traffic light control method based on fog calculation and reinforcement learning


    Additional title:

    一种基于雾计算和强化学习的交通灯控制方法


    Contributors:
    FAN XIUMEI (author) / AN MENGMENG (author)

    Publication date :

    2020-04-21


    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



    Deep reinforcement learning traffic light control method

    KONG YAN / LI YING / CHIH-CHAO YANG | European Patent Office | 2024

    Free access

    Traffic Light Control Using Reinforcement Learning

    Masfequier Rahman Swapno, S M / Nuruzzaman Nobel, SM / A C, Ramachandra et al. | IEEE | 2024


    Traffic light control method based on deep reinforcement learning and inverse reinforcement learning

    ZHANG YA / GU SHIYI / CHEN GUOXI | European Patent Office | 2023

    Free access

    Traffic light signal control method based on multi-agent reinforcement learning

    ZHAO SHENGJIE / DENG HAO / CHEN ZHI | European Patent Office | 2022

    Free access

    Deep Reinforcement Learning for Autonomous Traffic Light Control

    Garg, Deepeka / Chli, Maria / Vogiatzis, George | IEEE | 2018