SIA OpenIR  > 机器人学研究室
A framework for the robot skill learning using reinforcement learning
Wei YZ(魏英姿); Zhao MY(赵明扬)
Department机器人学研究室
Conference Name5th International Symposium on Instrumentation and Control Technology
Conference DateOctober 24-27, 2003
Conference PlaceBEIJING, China
Author of SourceNatl Nat Sci Fdn China, Chinese Soc Aeronaut & Astronaut, China Instrument & Control Soc, Chinese Soc Astronaut, Chinese Assoc Automat, SPIE, Beijing Univ Aeronaut & Astronaut, Univ Paderborn, Univ Essen, Hiroshima Shudo Univ, Pusan Natl Univ
Source PublicationFIFTH INTERNATIONAL SYMPOSIUM ON INSTRUMENTATION AND CONTROL TECHNOLOGY
PublisherSPIE-INT SOC OPTICAL ENGINEERING
Publication PlaceBELLINGHAM
2003
Pages910-914
Indexed ByEI ; CPCI(ISTP)
EI Accession number2004218170145
WOS IDWOS:000187954200193
Contribution Rank1
ISSN0277-786X
ISBN0-8194-5137-1
KeywordReinforcement Learning Robot Skill Learning Control Reward Function Cmac Neural Network
AbstractRobot acquiring skill is a process similar to human skill learning. Reinforcement learning (RL) is an on-line actor critic method for a robot to develop its skill. The reinforcement function has become the critical component for its effect of evaluating the action and guiding the learning process. We present an augmented reward function that provides a new way for RL controller to incorporate prior knowledge and experience into the RL controller. Also, the difference form of augmented reward function is considered carefully. The additional reward beyond conventional reward will provide more heuristic information for RL. In this paper, we present a strategy for the task of complex skill learning. Automatic robot shaping policy is to dissolve the complex skill into a hierarchical learning process. The new form of value function is introduced to attain smooth motion switching swiftly. We present a formal, but practical, framework for robot skill learning and also illustrate with an example the utility of method for learning skilled robot control on line.
Language英语
Citation statistics
Document Type会议论文
Identifierhttp://ir.sia.cn/handle/173321/8474
Collection机器人学研究室
Affiliation1.Shenyang Institute of Automation, Chinese Academy of Science, Shenyang, 110016, China
2.Shenyang Institute of Technology, Shenyang, 110168, China
3.Grad. Sch. the Chinese Acad. of Sci., Beijing, 100039, China
Recommended Citation
GB/T 7714
Wei YZ,Zhao MY. A framework for the robot skill learning using reinforcement learning[C]//Natl Nat Sci Fdn China, Chinese Soc Aeronaut & Astronaut, China Instrument & Control Soc, Chinese Soc Astronaut, Chinese Assoc Automat, SPIE, Beijing Univ Aeronaut & Astronaut, Univ Paderborn, Univ Essen, Hiroshima Shudo Univ, Pusan Natl Univ. BELLINGHAM:SPIE-INT SOC OPTICAL ENGINEERING,2003:910-914.
Files in This Item: Download All
File Name/Size DocType Version Access License
HYQW000044.pdf(238KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wei YZ(魏英姿)]'s Articles
[Zhao MY(赵明扬)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wei YZ(魏英姿)]'s Articles
[Zhao MY(赵明扬)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wei YZ(魏英姿)]'s Articles
[Zhao MY(赵明扬)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: HYQW000044.pdf
Format: Adobe PDF
This file does not support browsing at this time
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.