The invention provides an urban road network integrated control method based on coordination reinforcement learning, and the method comprises the steps: constructing a coordination decision-making structure composed of a double-actor network guided by different targets on the basis of constructing a network traffic environment comprising a human-driven vehicle and a network-connected automatic driving vehicle through traffic simulation; information sharing between double-actor networks is realized by adopting a centralized training and decentralized execution learning normal form and combining an attention mechanism, so that coordinated optimization control of vehicle guidance and traffic signal control is realized; and a comprehensive reward mechanism is adopted to dynamically allocate reward weights for different vehicles according to actual traffic demands so as to drive dynamic coordination among multiple agents.
本发明提供一种基于协调强化学习的城市路网集成控制方法,在交通仿真构建包含人类驾驶车辆和网联自动驾驶车辆的网联交通环境的基础上,构建由不同目标导向的双行动者网络组成的协调决策结构;通过采用集中式训练分散式执行学习范式,并结合注意力机制实现双行动者网络间的信息共享,以实现车辆诱导与交通信号控制的协调优化控制;并采用综合奖励机制,根据实际交通需求,为不同车辆动态分配奖励权重,以驱动多智能体之间的动态协调。
Urban road network integrated control method based on coordination reinforcement learning
一种基于协调强化学习的城市路网集成控制方法
2024-07-26
Patent
Electronic Resource
Chinese
IPC: | H04L TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION , Übertragung digitaler Information, z.B. Telegrafieverkehr / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Urban area road traffic coordination control method based on road network subarea division
European Patent Office | 2023
|Urban road automatic parking method and system based on reinforcement learning
European Patent Office | 2022
|Automobile stability coordination control method based on reinforcement learning
European Patent Office | 2025
|Bidirectional signal coordination control method for urban main road
European Patent Office | 2022
|Urban ecological road space driving behavior optimization method based on reinforcement learning
European Patent Office | 2024
|