何倩,仝武宁.基于深度学习的智能驾驶车辆路径仿真研究[J].中南民族大学学报自然科学版,2022,41(5):586-591
基于深度学习的智能驾驶车辆路径仿真研究
Research on intelligent driving vehicle routing simulation based on deep learning
  
DOI:10.12130/znmdzk.20220512
中文关键词: 智能驾驶  深度强化学习  DDPG算法  经验回放机制
英文关键词: intelligent driving  deep reinforcement learning  DDPG algorithm  experience playback mechanism
基金项目:陕西中医药大学科研启动基金资助项目(112-400231116)
作者单位
何倩 陕西中医药大学 计算机实验中心咸阳 712000 
仝武宁 陕西中医药大学 计算机实验中心咸阳 712000 
摘要点击次数: 105
全文下载次数: 83
中文摘要:
      基于深度强化学习技术研究了智能驾驶中的路径规划问题,且在虚拟环境下进行了模拟分析,对提出的路径规划算法性能做了验证研究.提出了一种改进的经验回放机制ERDDPG (Experience Replay Deep Deterministic Policy Gradient )算法,对经验样本通过优先经验回放机制处理而设置不同的优先级,高质量的经验样本被优先采样,这种模式下网络的训练效率显著提高.在仿真实验中,所提出的ERDDPG算法可完成智能驾驶的路径规划,学习效率较高,且智能车行驶的稳定性更好.
英文摘要:
      Based on the deep reinforcement learning technology, the path planning problem in intelligent driving is studied, and simulation analysis in the virtual environment is made to verify the performance of the proposed path planning algorithm. An improved experience replay mechanism ERDDPG (Experience Replay Deep Deterministic Policy Gradient) algorithm is proposed ,which sets different priorities for the experience samples through the priority experience replay mechanism, and the high-quality experience samples are firstly sampled. In this mode, the training efficiency of the network is significantly improved. In the simulation experiment, the ERDDPG algorithm proposed can complete the path planning of intelligent driving, with higher learning efficiency and better driving stability of intelligent vehicles.
查看全文   查看/发表评论  下载PDF阅读器
关闭