The invention belongs to the technical field of intelligent traffic, and particularly discloses an intersection intelligent network connection vehicle control method based on reinforcement learning, which comprises the following steps: step 1, obtaining the state quantity of an intelligent network connection vehicle at an intersection; 2, inputting the state quantity into the intersection intelligent network connection vehicle control network model, and obtaining a corresponding action and a preliminary reward; step 3, the steps 1-2 are cycled, and multiple groups of empirical data are acquired until the current vehicle leaves the intersection; 4, recording the number of circulation steps in the steps 1-2 and the passing time of the intelligent network connection vehicle, updating the initial reward in each group of experience data into a single-step reward, storing the updated experience data into an experience pool, and arranging serial numbers; 5, circulating the steps 1-4, and selecting empirical data from the empirical pool to update network parameters of the intersection intelligent network connection vehicle control network model; the invention provides a control method capable of controlling the intelligent networked vehicle to effectively perform active avoidance, obstacle avoidance, following and other actions at a complex intersection.

    本发明属于智能交通技术领域,具体公开一种基于强化学习的交叉口智能网联车控制方法,包括以下步骤:步骤1,获取交叉口的智能网联车的状态量;步骤2,将状态量输入交叉口智能网联车控制网络模型中,获取对应的动作和初步奖励;步骤3,循环步骤1‑2,获取多组经验数据,直到当前车辆驶离交叉口;步骤4,记录步骤1‑2的循环步数和智能网联车的通行时间,将每组经验数据中的初步奖励更新为单步奖励,将更新后的经验数据存入经验池中,并排列序号;步骤5,循环步骤1‑4,从经验池中选取经验数据更新交叉口智能网联车控制网络模型的网络参数;本发明提供了一种在复杂交叉口能够控制智能网联车有效进行主动避让、避障和跟随等动作的控制方法。


    Access

    Download


    Export, share and cite



    Title :

    Intersection intelligent network connection vehicle control method based on reinforcement learning


    Additional title:

    一种基于强化学习的交叉口智能网联车控制方法


    Contributors:
    WANG YOU (author) / SHI HENGLIANG (author) / CHEN HUAKUI (author) / MA LIN (author) / GAO HONGBO (author) / ZHOU YUNLONG (author) / MA CHICHENG (author) / QU YATING (author)

    Publication date :

    2024-01-30


    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung



    Intersection reinforcement learning signal control method for network connection vehicle detection state perception

    ZHANG JIAN / WANG HAN / WANG BO et al. | European Patent Office | 2022

    Free access

    Intersection management method based on intelligent network connection vehicle

    ZHAO SHENGJIE / XUE JINWEI / DENG HAO | European Patent Office | 2023

    Free access

    Intersection control method for intelligent network connection vehicles

    XIA FAN / ZHANG LUN | European Patent Office | 2020

    Free access

    Intelligent network connection vehicle on-ramp cooperative control method based on deep reinforcement learning

    XIE DONGFAN / LI JIAN / WANG ZIJIAN et al. | European Patent Office | 2023

    Free access

    Intersection and intelligent network connection vehicle control method and system and storage medium

    YU HAIYANG / WANG JIXIANG / REN YILONG et al. | European Patent Office | 2023

    Free access