The invention discloses a multi-agent reinforcement learning method for fair adaptive traffic signal control, and the method comprises the steps: carrying out the modeling of a fair adaptive traffic signal control problem as a constrained Markov game; the objective of the invention is to find an optimal joint strategy which can satisfy the cost constraint of each agent while maximizing the average expected return of all agents. On the basis of the constrained Markov game, using a confidence domain strategy optimization proxy function to optimize accumulated discount rewards and discount cost in average expectation return, and using a Lagrange multiplier method to convert a constrained problem into an unconstrained problem; and adopting vehicle fairness multi-agent approximation strategy optimization, and learning a cooperative control strategy for maximizing overall return of the unconstrained problem under the condition that waiting time constraint is satisfied. When traffic signal control is carried out on multiple intersections, efficiency and fairness are considered at the same time.

    本发明公开了一种面向公平自适应交通信号控制的多智能体强化学习方法,包括:将面向公平的自适应交通信号控制问题建模为约束马尔科夫博弈,目标是寻找一个最优的联合策略能在最大化所有智能体的平均期望回报的同时满足每个智能体的成本约束;在所述约束马尔可夫博弈的基础上,使用置信域策略优化代理函数优化平均期望回报中的累积折扣奖励和折扣成本,采用拉格朗日乘子法将有约束问题转化为无约束问题;采用车辆公平性多智能体近似策略优化,学习所述无约束问题在满足等待时间约束下最大化整体回报的协作控制策略。本发明在针对多交叉口进行交通信号控制时,同时兼顾了效率和公平性。


    Access

    Download


    Export, share and cite



    Title :

    Multi-agent reinforcement learning method for fair adaptive traffic signal control


    Additional title:

    面向公平自适应交通信号控制的多智能体强化学习方法


    Contributors:

    Publication date :

    2023-11-03


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Adaptive traffic signal control method based on multi-agent reinforcement learning

    ZHANG CHENGWEI / JIN SHAN / ZHENG KANGJIE | European Patent Office | 2021

    Free access

    Multi-agent reinforcement learning for integrated and networked adaptive traffic signal control

    EL-TANTAWY SAMAH / ABDULHAI BAHER | European Patent Office | 2017

    Free access

    MULTI-AGENT REINFORCEMENT LEARNING FOR INTEGRATED AND NETWORKED ADAPTIVE TRAFFIC SIGNAL CONTROL

    EL-TANTAWY SAMAH / ABDULHAI BAHER | European Patent Office | 2018

    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