The invention discloses an agent reinforcement learning method for controlling vehicle driving, and aims at training a control agent for vehicle driving in the field of unmanned driving. The intelligent body receives image data, radar data, motion data and the like observed in the advancing process of the vehicle in real time to serve as state input, calculates and outputs steering wheel rotation angle, accelerator control force and brake control force parameters to serve as action output, and obtains corresponding reward numerical value feedback according to the motion condition of the vehicle. The training target is to realize smooth driving of the vehicle on the premise of ensuring driving safety, keep a normal speed to the greatest extent, and avoid collision with other vehicles or objects. A plurality of factors such as driving speed, driving process loss, steering wheel angle/accelerator/brake operation matching degree and the like can influence reward feedback, and an intelligent agent needs to find a proper strategy to maximize a driving score. The invention further discloses a corresponding agent reinforcement learning device for controlling vehicle driving.

    本发明公开了一种用于控制车辆驾驶的智能体强化学习方法,针对无人驾驶领域,训练一个面向车辆驾驶的控制智能体。智能体实时接收车辆行进过程中观测到的影像数据、雷达数据、运动数据等作为状态输入,计算输出方向盘转角、油门控制力度、刹车控制力度参数作为动作输出,并根据车辆的运动情况获取相应的奖励数值反馈,训练目标是实现车辆在保证驾驶安全的前提下顺利行驶,同时最大程度上保持正常速度,并避免与其它车辆或物体产生碰撞。驾驶速度、行驶过程损耗、方向盘转角/油门/刹车操作契合程度等多种因素都会影响奖励反馈,智能体需要找到一个合适的策略使得行驶得分最大化。本发明还公开了相应的用于控制车辆驾驶的智能体强化学习装置。


    Access

    Download


    Export, share and cite



    Title :

    Intelligent agent reinforcement learning method and device for controlling vehicle driving


    Additional title:

    用于控制车辆驾驶的智能体强化学习方法与装置


    Contributors:
    GAO ZIWEN (author) / LIU JUNTAO (author) / WANG YUANBIN (author) / WANG ZHENJIE (author) / REN WEILIN (author)

    Publication date :

    2023-05-30


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung



    Automatic driving vehicle control method based on multi-agent reinforcement learning

    MA HONGBIN / LIU PING / JIN YING et al. | European Patent Office | 2023

    Free access

    Intelligent driving vehicle autonomous parking system based on deep reinforcement learning

    CAO XIANGSHANG / LI YONG / ZHAO WEI | European Patent Office | 2024

    Free access

    Method and device for controlling vehicle driving and intelligent vehicle

    HE MIN / LIU QI / XIONG JUN et al. | European Patent Office | 2023

    Free access

    Automatic driving vehicle high-speed ramp intelligent afflux method based on reinforcement learning

    HU XIAOSONG / XIAO WENXUAN / YANG YUYOU et al. | European Patent Office | 2023

    Free access

    VEHICLE DRIVING LEARNING SYSTEM USING REINFORCEMENT LEARNING

    SONG MYUNGHO / LEE JOONGHEE / KIM JUYOUNG et al. | European Patent Office | 2022

    Free access