The invention discloses a lane keeping method based on deep reinforcement learning, and the method comprises the steps: collecting environment-agent state observation values, namely environment state data and vehicle state data, carrying out the preprocessing of the data, transmitting the data into a network model, and initializing a network. And the network learns a lane to keep the driving task in a reinforcement learning mode based on interaction data, the network is continuously updated during the period until the model converges, and then optimal strategy output control of task execution is performed based on the converged network. According to the lane keeping method based on deep reinforcement learning provided by the invention, self-learning of the lane keeping task of the trolley is realized, a traditional lane detection-processing-control process is simplified, and end-to-end command output control is realized.

    本发明公开了基于深度强化学习的车道保持方法,包括采集环境‑智能体状态观测值,即环境状态数据和车辆状态数据并进行数据的预处理后用以传入网络模型和初始化网络。网络基于交互数据以强化学习的方式学习车道保持这一驾驶任务,期间不断更新网络直至模型收敛,然后基于收敛后的网络进行任务执行的最优策略输出控制。本发明提供的基于深度强化学习的车道保持方法,实现了小车车道保持任务的自学习,简化了传统的车道检测‑处理‑控制流程,实现了端到端的命令输出控制。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Lane keeping method based on deep reinforcement learning


    Weitere Titelangaben:

    基于深度强化学习的车道保持方法


    Beteiligte:
    YANG DONGLI (Autor:in) / ZHANG JUNYANG (Autor:in) / LUO KE (Autor:in) / FENG MINGJIE (Autor:in)

    Erscheinungsdatum :

    2021-12-28


    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06K Erkennen von Daten , RECOGNITION OF DATA / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Lane keeping decision-making method based on deep reinforcement learning

    WANG PENGWEI / ZHOU HENGHENG / GAO SONG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Lane Keeping Assist for an Autonomous Vehicle Based on Deep Reinforcement Learning

    Wang, Qun / Zhuang, Weichao / Wang, Liangmo et al. | British Library Conference Proceedings | 2020


    Lane Keeping Assist for an Autonomous Vehicle Based on Deep Reinforcement Learning

    Wang, Qun / Zhuang, Weichao / Wang, Liangmo et al. | British Library Conference Proceedings | 2020


    Lane Keeping Assist for an Autonomous Vehicle Based on Deep Reinforcement Learning

    Ju, Fei / Wang, Liangmo / Wang, Qun et al. | SAE Technical Papers | 2020


    Personalized lane keeping assistance method based on deep learning

    DU XIANGHUA / QIN SONG / LIU GANG | Europäisches Patentamt | 2022

    Freier Zugriff