The invention discloses a multi-agent intelligent signal lamp road network control method based on a deep reinforcement learning strategy. According to the method, the overall behavior space is reduced according to the idea of multiple agents, and the model training efficiency is improved by considering global and local reward functions. Based on the position information of each signal lamp and a multi-agent method, an optimal strategy is determined by using a reinforcement learning algorithm, and the signal lamps are allowed to cooperatively work to seek a globally optimal solution, so that an optimal control strategy of the road network signal lamps is realized. A multi-agent method is adopted to control the road network signal lamp, and when other problems are introduced, the difficulty of model optimization can be reduced.
本发明公开了一种基于深度强化学习策略的多智能体智能信号灯路网控制方法。本发明根据多智能体的思想降低整体的行为空间,通过考虑全局和局部的奖赏函数,提升了模型训练效率。本发明基于每个信号灯的位置信息和多智能体方法,使用强化学习的算法确定最优策略,并允许各信号灯之间协同工作以寻求全局最优解,实现路网信号灯的最优控制策略。本发明采用多智能体方法进行路网信号灯的控制,在引入其他问题时,能够减轻模型优化的困难。
Multi-agent intelligent signal lamp road network control method based on deep reinforcement learning strategy
基于深度强化学习策略的多智能体智能信号灯路网控制方法
2021-04-23
Patent
Electronic Resource
Chinese
Signal lamp intelligent control method based on deep reinforcement learning
European Patent Office | 2024
|Multi-agent road traffic signal control method based on deep reinforcement learning algorithm
European Patent Office | 2023
|Multi-agent reinforcement learning-based multi-traffic signal lamp control method and system
European Patent Office | 2024
|European Patent Office | 2024
|Traffic signal control method based on multi-agent deep reinforcement learning
European Patent Office | 2025
|