Acta Scientiarum Naturalium Universitatis Pekinensis ›› 2023, Vol. 59 ›› Issue (6): 1069-1086.DOI: 10.13209/j.0479-8023.2023.086

Previous Articles     Next Articles

A Review of Robot Learning

QU Weiming1,*, LIU Tianlin1,*, LIN Weikai1, LUO Dingsheng1,2,†   

  1. 1. School of Intelligence Science and Technology, Peking University, Beijing 100871 2. PKU-Wuhan Institute for Artificial Intelligence, Wuhan 430073
  • Received:2022-11-28 Revised:2023-04-25 Online:2023-11-20 Published:2023-11-20
  • Contact: LUO Dingsheng, E-mail: dsluo(at)pku.edu.cn

机器人学习方法综述

曲威名1,*, 刘天林1,*, 林惟凯1, 罗定生1,2,†   

  1. 1. 北京大学智能学院, 北京 100871 2. 北京大学武汉人工智能研究院, 武汉 430073
  • 通讯作者: 罗定生, E-mail: dsluo(at)pku.edu.cn
  • 基金资助:
    国家自然科学基金(62176004, U1713217)、东湖高新区国家智能社会治理实验综合基地项目、北京大学新工科专项、北京大学–辛巴科技项目和北京大学高性能计算平台资助

Abstract:

The basic concepts and core issues related to robot learning are introduced and discussed, and the relevant researches are summarized and analyzed. Through comparing the relevant methods and recent progress, the authors classify the methods of robot learning into four categories based on data types and learning methods, namely reinforcement learning approach, imitation learning approach, transfer learning approach and developmental learning approach. Finally, current challenges and future trends in robot learning are listed.

Key words: robot learning, reinforcement learning, imitation learning, transfer learning, developmental learning

摘要:

介绍与机器人学习有关的基本概念与核心问题, 梳理机器人学习的相关方法和最新进展。依据数据类型, 将机器人学习的方法分为基于强化学习的方法、基于模仿学习的方法、基于迁移学习的方法和基于发展学习的方法, 并对相关研究进行总结和分析, 探讨机器人学习领域目前存在的挑战和未来发展趋势。

关键词: 机器人学习, 强化学习, 模仿学习, 迁移学习, 发展学习