Show simple item record

dc.contributor.authorKober, J.
dc.contributor.authorWilhelm, A.
dc.contributor.authorÖztop, Erhan
dc.contributor.authorPeters, J.
dc.date.accessioned2014-05-31T12:31:58Z
dc.date.available2014-05-31T12:31:58Z
dc.date.issued2012-11
dc.identifier.issn1573-7527
dc.identifier.urihttp://hdl.handle.net/10679/364
dc.identifier.urihttp://link.springer.com/article/10.1007%2Fs10514-012-9290-3
dc.description.abstractHumans manage to adapt learned movements very quickly to new situations by generalizing learned behaviors from similar situations. In contrast, robots currently often need to re-learn the complete movement. In this paper, we propose a method that learns to generalize parametrized motor plans by adapting a small set of global parameters, called meta-parameters. We employ reinforcement learning to learn the required meta-parameters to deal with the current situation, described by states. We introduce an appropriate reinforcement learning algorithm based on a kernelized version of the reward-weighted regression. To show its feasibility, we evaluate this algorithm on a toy example and compare it to several previous approaches. Subsequently, we apply the approach to three robot tasks, i.e., the generalization of throwing movements in darts, of hitting movements in table tennis, and of throwing balls where the tasks are learned on several different real physical robots, i.e., a Barrett WAM, a BioRob, the JST-ICORP/SARCOS CBi and a Kuka KR 6.en_US
dc.description.sponsorshipEuropean Communityen_US
dc.language.isoengen_US
dc.publisherSpringer Science+Business Mediaen_US
dc.relationinfo:eurepo/grantAgreement/EC/FP7/270327en_US
dc.relation.ispartofAutonomous Robots
dc.rightsopenAccess
dc.titleReinforcement learning to adjust parametrized motor primitives to new situationsen_US
dc.typeArticleen_US
dc.peerreviewedyesen_US
dc.publicationstatuspublisheden_US
dc.contributor.departmentÖzyeğin University
dc.contributor.authorID45227
dc.contributor.ozuauthorÖztop, Erhan
dc.identifier.volume33
dc.identifier.issue4
dc.identifier.startpage361
dc.identifier.endpage379
dc.identifier.wosWOS:000307766800002
dc.identifier.doi10.1007/s10514-012-9290-3
dc.subject.keywordsSkill learningen_US
dc.subject.keywordsMotor primitivesen_US
dc.subject.keywordsReinforcement learningen_US
dc.subject.keywordsMeta-parametersen_US
dc.subject.keywordsPolicy learningen_US
dc.identifier.scopusSCOPUS:2-s2.0-84868358933
dc.contributor.authorMale1


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record


Share this page