Efficient Insertion Control for Precision Assembly Based on Demonstration Learning and Reinforcement Learning

doi:10.1109/TII.2020.3020065

CORC > 自动化研究所 > 中国科学院自动化研究所 > 精密感知与控制研究中心 > 精密感知与控制

	Efficient Insertion Control for Precision Assembly Based on Demonstration Learning and Reinforcement Learning
	Ma, Yanqin 1,2; Xu, De 2; Qin, Fangbo 2
刊名	IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
	2021-07-01
卷号	17 期号:7 页码:4492-4502
关键词	Learning (artificial intelligence) Task analysis Learning systems Gaussian distribution Informatics Automation Data models Demonstration learning insertion policy learning multiple peg-in-hole insertion precision assembly reinforcement learning
ISSN号	1551-3203
DOI	10.1109/TII.2020.3020065
通讯作者	Xu, De(de.xu@ia.ac.cn)
英文摘要	Multiple peg-in-hole insertion control is one of the challenging tasks in precision assembly for its complex contact dynamics. In this article, an insertion policy learning method is proposed for multiple peg-in-hole precision assembly. The insertion policy learning process is separated into two phases: initial policy learning and residual policy learning. In initial policy learning, a state-to-action policy mapping model based on the Gaussian mixture model (GMM) is established. And Gaussian mixture regression (GMR) is used to generalize the policy reuse. In residual policy learning, a reinforcement learning method named normalized advantage function (NAF) is employed to refine the insertion policy via agent's exploration in the insertion environment. Moreover, an adaptive action exploration (AAE) strategy is designed to improve the performance of exploration, and the prioritized experience replay strategy is introduced to make the residual policy learning from historical experience more efficient. Besides, the hierarchical reward function is designed considering the contact dynamics as well as the efficiency and safety of precision insertion. Finally, comprehensive experiments are conducted to validate the effectiveness of the proposed insertion policy learning method.
资助项目	National Natural Science Foundation of China[61873266] ; National Natural Science Foundation of China[61733004]
WOS研究方向	Automation & Control Systems ; Computer Science ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000638402700007
资助机构	National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/44260]
专题	精密感知与控制研究中心_精密感知与控制
通讯作者	Xu, De
作者单位	1.Nanjing Vocat Univ Ind Technol, Sch Comp & Software, Nanjing 210023, Peoples R China 2.Chinese Acad Sci, Res Ctr Precis Sensing & Control, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Ma, Yanqin,Xu, De,Qin, Fangbo. Efficient Insertion Control for Precision Assembly Based on Demonstration Learning and Reinforcement Learning[J]. IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS,2021,17(7):4492-4502.
APA	Ma, Yanqin,Xu, De,&Qin, Fangbo.(2021).Efficient Insertion Control for Precision Assembly Based on Demonstration Learning and Reinforcement Learning.IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS,17(7),4492-4502.
MLA	Ma, Yanqin,et al."Efficient Insertion Control for Precision Assembly Based on Demonstration Learning and Reinforcement Learning".IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 17.7(2021):4492-4502.