Reinforcement Learning to adjust Robot Movements to New Situations

Kober, J; Oztop, E; Peters, J

doi:10.15607/RSS.2010.VI.005

Local TagsRelease HistoryDetailsSummary

Reinforcement Learning to adjust Robot Movements to New Situations

Kober, J., Oztop, E., & Peters, J. (2011). Reinforcement Learning to adjust Robot Movements to New Situations. Robotics: Science and Systems VI, 33-40.

Item is Released

show all hide all

Basic

show hide

Item Permalink: http://hdl.handle.net/11858/00-001M-0000-0013-BA64-3 Version Permalink: http://hdl.handle.net/21.11116/0000-0002-061F-6

Genre: Conference Paper

Files

show Files

Locators

show

hide

Locator:
http://www.roboticsproceedings.org/rss06/p05.pdf (Publisher version)

Description:
-

Creators

show

hide

Creators:
Kober, J^{1, 2}, Author
Oztop, E, Author
Peters, J^{1, 2}, Author

Affiliations:
1Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497795
2Max Planck Institute for Biological Cybernetics, Max Planck Society, Spemannstrasse 38, 72076 Tübingen, DE, ou_1497794

Content

show

hide

Free keywords: -

Abstract: Many complex robot motor skills can be represented using elementary movements, and there exist efficient techniques for learning parametrized motor plans using demonstrations and self-improvement. However, in many cases, the robot currently needs to learn a new elementary movement even if a parametrized motor plan exists that covers a similar, related situation. Clearly, a method is needed that modulates the elementary movement through the meta-parameters of its representation. In this paper, we show how to learn such mappings from circumstances to meta-parameters using reinforcement learning.We introduce an appropriate reinforcement learning algorithm based on a kernelized version of the reward-weighted regression. We compare this algorithm to several previous methods on a toy example and show that it performs well in comparison to standard algorithms. Subsequently, we show two robot applications of the presented setup; i.e., the generalization of throwing movements in darts, and of hitting movements in table tennis. We show that both tasks can be learned successfully using simulated and real robots.

Details

show

hide

Language(s):

Dates: Published in Print: 2011-09

Publication Status: Published in print

Pages: -

Publishing info: -

Table of Contents: -

Rev. Type: -

Identifiers: BibTex Citekey: 6438
DOI: 10.15607/RSS.2010.VI.005

Degree: -

Event

show

hide

Title: 2010 Robotics: Science and Systems Conference (RSS 2010)

Place of Event: Zaragoza, Spain

Start-/End Date: 2011-06-27 - 2011-06-30

Legal Case

show

Project information

show

Source 1

show

hide

Title: Robotics: Science and Systems VI

Source Genre: Journal

Creator(s):
Matsuoka, Y, Editor
Durrant-Whyte, HF, Editor
Neira, J, Editor

Affiliations:
-

Publ. Info: Cambridge, MA, USA : MIT Press

Pages: - Volume / Issue: - Sequence Number: - Start / End Page: 33 - 40 Identifier: ISBN: 978-0-262-51681-5