The invention discloses a driver longitudinal car-following behavior model construction method based on deep reinforcement learning, and belongs to the field of automobile intelligent safety and automatic driving. The method comprises the steps of based on the actual road working condition of China, collecting vehicle state information and surrounding environment information, meeting the road characteristics of China, of a driver in the vehicle driving process, counting and analyzing the collected data, and giving behavior characteristics and influence factors of the driver in the car following driving process; determining reference information representing actions taken by the driver at a certain moment, and establishing a mathematical model for describing the iterative relationship of the driver car-following behavior state; designing a neural network structure of the driver longitudinal car-following behavior model based on the competitive Q network architecture; designing a driverlongitudinal car-following behavior learning process of a neural network based on the competitive Q network architecture; and designing a training method of the driver longitudinal vehicle following behavior model based on deep reinforcement learning. The car following behavior characteristics of the driver under different working conditions can be accurately described, and the reproduction capability of the car following behavior of the driver is achieved.

    基于深度强化学习的驾驶员纵向跟车行为模型构建方法,属于汽车智能安全与自动驾驶领域。基于中国实际道路工况,采集符合中国道路特征的驾驶员驾驶车辆行驶过程中的车辆状态信息和周围环境信息,统计分析采集的数据,给出驾驶员跟车行驶过程的行为特性及其影响因素。确定表征驾驶员在某个时刻所采取动作的基准信息,建立描述驾驶员跟车行为状态迭代关系的数学模型。设计基于竞争Q网络构架的驾驶员纵向跟车行为模型的神经网络结构。设计基于竞争Q网络构架的神经网络的驾驶员纵向跟车行为学习流程。设计基于深度强化学习的驾驶员纵向跟车行为模型的训练方法。可准确地描述不同工况下驾驶员的跟车行为特性,实现对驾驶员跟车行为的复现能力。


    Access

    Download


    Export, share and cite



    Title :

    Driver longitudinal car-following behavior model construction method based on deep reinforcement learning


    Additional title:

    基于深度强化学习的驾驶员纵向跟车行为模型构建方法


    Contributors:
    GUO JINGHUA (author) / LI WENCHANG (author) / WANG JINGYAO (author) / WANG BAN (author) / XIAO BAOPING (author)

    Publication date :

    2021-01-08


    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 / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen





    Construction method of deep reinforcement learning vehicle following model fused with driving style

    WANG KEYIN / SHI ZHEN / ZHANG JIANHUI et al. | European Patent Office | 2022

    Free access


    DEEP REINFORCEMENT LEARNING FOR A GENERAL FRAMEWORK FOR MODEL-BASED LONGITUDINAL CONTROL

    PATHAK SHASHANK / BAG SUVAM / NADKARNI VIJAY JAYANT | European Patent Office | 2020

    Free access