According to the vehicle formation method provided by the invention, a plurality of unmanned vehicles can cooperatively run on a road, air resistance is reduced, and fuel consumption is reduced. The state and action space of the vehicle formation and the traffic light intelligent agent are defined, so that the vehicle formation and the traffic light intelligent agent can better adapt to traffic control requirements, and the current environment state is observed. A vehicle formation traffic light intelligent agent cooperative reward method is designed, and intelligent agents are encouraged to cooperatively control traffic light signals so as to optimize traffic fluency and reduce fuel consumption. A vehicle formation agent selection mechanism is provided, the training time can be shortened, rapid convergence of the algorithm can be promoted, and the efficiency of practical application is improved. Cooperative control of the vehicle formation and the traffic lights at the intersection is realized, so that energy consumption is reduced, intersection congestion is relieved, and the method is of great significance to improvement of urban traffic. A training method based on a PPO reinforcement learning model is designed to effectively train an intelligent agent and support the intelligent agent to make an intelligent decision in a complex traffic environment.

    本发明提出了一种车辆编队方法,使多辆无人车能够在道路上协同行驶,减少空气阻力,降低燃料消耗。定义了车辆编队和交通灯智能体的状态和动作空间,使其能够更好地适应交通控制的需求,观测当前环境状态。设计了一种车辆编队交通灯智能体协同的奖励方法,鼓励智能体协同控制交通灯信号,以优化交通流畅度,降低燃料消耗。提出了一种车辆编队智能体选择机制,可以减少训练时间和促进算法的快速收敛,提高了实际应用的效率。实现了交叉路口车辆编队与交通灯的协同控制,以降低能源消耗,减轻交叉口拥堵,对城市交通的改善具有重要意义。设计了一种基于PPO强化学习模型的训练方法,以有效训练智能体,并支持其在复杂交通环境中做出智能决策。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Vehicle formation and signal lamp cooperative control method and system based on multi-agent reinforcement learning


    Weitere Titelangaben:

    一种基于多智能体强化学习的车辆编队与信号灯协同控制方法及系统


    Beteiligte:
    TANG HAICHUAN (Autor:in) / GONG MING (Autor:in) / FU YUNXIAO (Autor:in) / JI HONGLEI (Autor:in) / SUN YIFAN (Autor:in)

    Erscheinungsdatum :

    2024-03-08


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Signal lamp and route cooperative control method based on offline reinforcement learning

    OUYANG YAJIE / ZHAO KUO | Europäisches Patentamt | 2024

    Freier Zugriff

    Multi-Agent Deep Reinforcement Learning for Decentralized Cooperative Traffic Signal Control

    Zhao, Yang / Hu, Jian-Ming / Gao, Ming-Yang et al. | ASCE | 2020


    Bus priority traffic signal cooperative control method based on multi-agent deep reinforcement learning

    WANG CONGYU / WANG CHONG / LI LEQI et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Multi-agent intelligent signal lamp road network control method based on deep reinforcement learning strategy

    LIU XUELI / YIN BAOCAI / GAO WEN | Europäisches Patentamt | 2021

    Freier Zugriff

    Multi-agent Cooperative Search based on Reinforcement Learning

    Sun, Yinjiang / Zhang, Rui / Liang, Wenbao et al. | IEEE | 2020