基于深度强化学习的机器人运动控制研究进展
作者:
作者单位:

1.贵州大学机械工程学院;2.贵州大学现代制造技术教育部重点实验室;3.公共大数据国家重点实验室(贵州大学)

作者简介:

通讯作者:

中图分类号:

TP242

基金项目:

国家重点研发计划(2018AAA0101801);国家自然科学基金项目(51475097,91746116);工信部资助项目(No.工信部联装[2016]213号);贵州省科技计划项目(黔科合人才[2015]4011);贵州省重点实验室建设项目(黔科合平台人才[2016]5103);贵州大学培育项目(贵大培育[2019]22号)。


Research Progress of Robot Motion Control Based on Deep Reinforcement Learning
Author:
Affiliation:

1.College of Mechanical Engineering, Guizhou University;2.Key Laboratory of Advanced Manufacturing Technology of Ministry of Education, Guizhou University;3.State Key Laboratory of Public Big Data (Guizhou University), Guizhou University

Fund Project:

The National Key Research and Development Program (2018AAA0101801); The National Natural Science Foundation of China under Grant (51475097, 91746116); The Ministry of Industry and Information Technology of the People’s Republic of China ([2016]213); Science and Technology Project of Guizhou Province under Grant Nos. (Talents[2015]4011); key Laboratory Construction project of Guizhou Province (Talents[2016]5103); and Cultivation Project of Guizhou University(Cultivation [2019]No.22 of Guizhou University)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    复杂未知环境下智能感知与自动控制是目前机器人在控制领域的研究热点之一,而新一代人工智能为其实现智能自动化赋予了可能.近年来,在高维连续状态-动作空间中,尝试运用深度强化学习进行机器人运动控制的新兴方法受到了相关研究人员的关注.本篇综述首先回顾了深度强化学习的兴起与发展,将用于机器人运动控制的深度强化学习算法分为基于值函数和策略梯度2类,并对各自典型算法及其特点进行了详细介绍;其次,针对仿真至现实之前的学习过程,简要介绍了5种常用于深度强化学习的机器人运动控制仿真平台;然后根据研究类型的不同,综述了目前基于深度强化学习的机器人运动控制方法在自主导航、物体抓取、步态控制、人机协作以及群体协同等5个方面的研究进展.最后,对其未来所面临的挑战以及发展趋势进行了总结与展望.

    Abstract:

    Intelligent perception and automatic control in a complex unknown environment is one of the current research hotspots of robots in the field of control, and a new generation of artificial intelligence makes it possible to realize intelligent automation. In recent years, the new method of robot control using deep reinforcement learning in high-dimensional continuous state-action space has attracted the attention of relevant researchers. In this survey, the rise and development of deep reinforcement learning are first reviewed. Then, the deep reinforcement learning algorithms for robot motion control are classified into two categories: value-based functions and policy gradients, and their typical algorithms and their related features are detailly described. Next, for the learning process before simulation to reality, five kinds of simulation platform for robot motion control are briefly introduced, which are often used for deep reinforcement learning. Moreover, according to different types of research, the research progress of the deep reinforcement learning approach to robot motion control is expounded in five aspects, including autonomous navigation, object grasping, gait control, human-robot collaborative and multi-robot cooperation. Finally, the future challenges and development trends are summarized and anticipated.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-10-08
  • 最后修改日期:2021-02-19
  • 录用日期:2021-03-03
  • 在线发布日期:
  • 出版日期: