The invention discloses an automatic driving decision-making system based on deep reinforcement learning. The automatic driving decision-making system comprises a vehicle sensing module, a generator SAC learning module and a discriminator judgment module, the vehicle sensing module obtains visual data and a driving state of a vehicle based on a sensor of the vehicle, and obtains sensing features of the vehicle through a convolutional neural network; the generator SAC learning module learns the perceptual characteristics of the vehicle to obtain the action of the vehicle; and the discriminator judgment module scores by judging the difference between the action obtained by the generator SAC learning module and the expert data to obtain a discriminator reward. A GAIL structure is utilized, the score of a discriminator on a generator is used as a discriminator reward, and the environment reward and the generative adversarial imitation learning score are linearly weighted to obtain a reward function of the model. And then, through the learning of the GAIL on expert data, expert knowledge is used for guiding the training of the intelligent agent, so that the training efficiency of the model is improved.

    本发明公开一种基于深度强化学习的自动驾驶决策系统,包括:车辆感知模块、生成器SAC学习模块及鉴别器判断模块;车辆感知模块基于车辆的传感器获得车辆的视觉数据和驾驶状态,通过卷积神经网络得到车辆的感知特征;生成器SAC学习模块通过对车辆的感知特征进行学习,得到车辆的动作;鉴别器判断模块通过判断生成器SAC学习模块得到的动作与专家数据的差异进行打分,得到鉴别器奖励。利用GAIL的结构,将鉴别器对生成器的打分作为鉴别器奖励,将环境奖励和生成对抗式模仿学习评分线性加权得到模型的奖励函数。然后通过对GAIL对专家数据的学习,用专家知识指导智能体的训练提高了模型的训练效率。


    Access

    Download


    Export, share and cite



    Title :

    Automatic driving decision-making system based on deep reinforcement learning


    Additional title:

    基于深度强化学习的自动驾驶决策系统


    Contributors:
    ZHENG XIAOYAO (author) / YAO QINGHE (author) / ZHANG JIANPENG (author) / ZHAO SIQI (author) / TENG LI (author) / LUO YONGLONG (author)

    Publication date :

    2024-05-03


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



    Automatic driving decision-making method based on deep reinforcement learning

    LIU CHENGQI / LIU SHAOWEIHUA / ZHANG YUJIE et al. | European Patent Office | 2024

    Free access

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

    YANG MINGZHU / LIU XIANGWEI / LI ZHUOLUO | European Patent Office | 2020

    Free access

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

    ZOU JUNYI / WANG HAIXIN | European Patent Office | 2024

    Free access

    Automatic driving lane keeping decision-making method based on deep reinforcement learning

    LIU HAOLIN / ZHANG XIAOHUI / BAI WENQI et al. | European Patent Office | 2025

    Free access

    Hybrid automatic driving lane changing decision-making method based on deep reinforcement learning

    LIANG SHANSHAN / ZHANG JIE / CHEN CHONGCHONG et al. | European Patent Office | 2024

    Free access