The invention discloses a cascade PID automatic driving longitudinal control method based on deep reinforcement learning. The cascade PID automatic driving longitudinal control method comprises the following steps: constructing a control instruction mapping mechanism; vehicle state information is obtained, and the acceleration compensation amount is obtained by constructing a cascade PID controller; optimizing cascade PID controller parameters based on deep reinforcement learning; obtaining an optimized Actor network parameter through a total reward function; and generating a target acceleration based on the optimized cascade PID controller parameters and executing control. According to the method, the formed cascade PID controller is combined with the deep reinforcement learning algorithm, so that the cascade PID controller can dynamically adjust parameters according to real-time driving feedback and environment change, the self-adaptive capability and robustness of the system are remarkably improved, and when the automatic driving system faces complex and changeful road conditions, the system can be effectively controlled. And more intelligent driving control can be realized.

    本发明公开了一种基于深度强化学习的级联PID自动驾驶纵向控制方法,所述级联PID自动驾驶纵向控制方法包括以下步骤:构建控制指令映射机制;获取车辆状态信息,通过构建级联PID控制器,获得加速补偿量;基于深度强化学习,优化的级联PID控制器参数;通过总奖励函数,获得优化后的Actor网络参数;基于优化的级联PID控制器参数生成目标加速度并执行控制。本方法通过构成的级联PID控制器结合深度强化学习算法,使级联PID控制器能够根据实时驾驶反馈和环境变化动态调整参数,显著提高系统的自适应能力和鲁棒性,在自动驾驶系统面对复杂和变化多端的路况时,能够实现更加智能化的驾驶控制。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Cascade PID automatic driving longitudinal control method based on deep reinforcement learning


    Weitere Titelangaben:

    一种基于深度强化学习的级联PID自动驾驶纵向控制方法


    Beteiligte:
    MENG MENG (Autor:in) / ZHOU XUN (Autor:in) / CHEN GUISONG (Autor:in)

    Erscheinungsdatum :

    16.05.2025


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    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



    Automatic driving transverse and longitudinal integrated decision-making method based on deep reinforcement learning

    IM MYOUNG-YOON / WANG JUAN / ZHOU JUNJIE et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Automatic driving decision planning method based on deep reinforcement learning and deep learning

    YANG LU / ZHANG HAO / TAN YANSONG et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Automatic driving decision-making method based on deep reinforcement learning

    LIU CHENGQI / LIU SHAOWEIHUA / ZHANG YUJIE et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Automatic driving behavior decision-making method based on deep reinforcement learning

    ZOU JUNYI / WANG HAIXIN | Europäisches Patentamt | 2024

    Freier Zugriff

    Automatic driving behavior decision-making method based on deep reinforcement learning

    YANG MINGZHU / LIU XIANGWEI / LI ZHUOLUO | Europäisches Patentamt | 2020

    Freier Zugriff