Learned parametrized dynamic movement primitives with shared synergies for controlling robotic and musculoskeletal systems

Research output: Contribution to journalArticleResearchpeer-review

Standard

Learned parametrized dynamic movement primitives with shared synergies for controlling robotic and musculoskeletal systems. / Rückert, Elmar; d'Avella, Andrea.
In: Frontiers in computational neuroscience, Vol. 7.2013, No. October, 17.10.2013.

Research output: Contribution to journalArticleResearchpeer-review

Bibtex - Download

@article{073182d5204142f696de6c3ae2d4b6d0,
title = "Learned parametrized dynamic movement primitives with shared synergies for controlling robotic and musculoskeletal systems",
abstract = "Asalientfeatureofhumanmotorskilllearningistheabilitytoexploitsimilaritiesacrossrelatedtasks.Inbiologicalmotorcontrol,ithasbeenhypothesizedthatmusclesynergies,coherentactivationsofgroupsofmuscles,allowforexploitingsharedknowledge.Recentstudieshaveshownthatarichsetofcomplexmotorskillscanbegeneratedbyacombinationofasmallnumberofmusclesynergies.Inrobotics,dynamicmovementprimitivesarecommonlyusedformotorskilllearning.Thismachinelearningapproachimplementsastableattractorsystemthatfacilitateslearninganditcanbeusedinhigh-dimensionalcontinuousspaces.However,itdoesnotallowforreusingsharedknowledge,i.e.,foreachtaskanindividualsetofparametershastobelearned.Weproposeanovelmovementprimitiverepresentationthatemploysparametrizedbasisfunctions,whichcombinesthebenefitsofmusclesynergiesanddynamicmovementprimitives.Foreachtaskasuperpositionofsynergiesmodulatesastableattractorsystem.Thisapproachleadstoacompactrepresentationofmultiplemotorskillsandatthesametimeenablesefficientlearninginhigh-dimensionalcontinuoussystems.Themovementrepresentationsupportsdiscreteandrhythmicmovementsandinparticularincludesthedynamicmovementprimitiveapproachasaspecialcase.Wedemonstratethefeasibilityofthemovementrepresentationinthreemulti-tasklearningsimulatedscenarios.First,thecharacteristicsoftheproposedrepresentationareillustratedinapoint-masstask.Second,incomplexhumanoidwalkingexperiments,multiplewalkingpatternswithdifferentstepheightsarelearnedrobustlyandefficiently.Finally,inamulti-directionalreachingtasksimulatedwithamusculoskeletalmodelofthehumanarm,weshowhowtheproposedmovementprimitivescanbeusedtolearnappropriatemuscleexcitationpatternsandtogeneralizeeffectivelytonewreachingskills.",
author = "Elmar R{\"u}ckert and Andrea d'Avella",
year = "2013",
month = oct,
day = "17",
doi = "10.3389/fncom.2013.00138",
language = "Undefined/Unknown",
volume = "7.2013",
journal = "Frontiers in computational neuroscience",
issn = "1662-5188",
number = "October",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Learned parametrized dynamic movement primitives with shared synergies for controlling robotic and musculoskeletal systems

AU - Rückert, Elmar

AU - d'Avella, Andrea

PY - 2013/10/17

Y1 - 2013/10/17

N2 - Asalientfeatureofhumanmotorskilllearningistheabilitytoexploitsimilaritiesacrossrelatedtasks.Inbiologicalmotorcontrol,ithasbeenhypothesizedthatmusclesynergies,coherentactivationsofgroupsofmuscles,allowforexploitingsharedknowledge.Recentstudieshaveshownthatarichsetofcomplexmotorskillscanbegeneratedbyacombinationofasmallnumberofmusclesynergies.Inrobotics,dynamicmovementprimitivesarecommonlyusedformotorskilllearning.Thismachinelearningapproachimplementsastableattractorsystemthatfacilitateslearninganditcanbeusedinhigh-dimensionalcontinuousspaces.However,itdoesnotallowforreusingsharedknowledge,i.e.,foreachtaskanindividualsetofparametershastobelearned.Weproposeanovelmovementprimitiverepresentationthatemploysparametrizedbasisfunctions,whichcombinesthebenefitsofmusclesynergiesanddynamicmovementprimitives.Foreachtaskasuperpositionofsynergiesmodulatesastableattractorsystem.Thisapproachleadstoacompactrepresentationofmultiplemotorskillsandatthesametimeenablesefficientlearninginhigh-dimensionalcontinuoussystems.Themovementrepresentationsupportsdiscreteandrhythmicmovementsandinparticularincludesthedynamicmovementprimitiveapproachasaspecialcase.Wedemonstratethefeasibilityofthemovementrepresentationinthreemulti-tasklearningsimulatedscenarios.First,thecharacteristicsoftheproposedrepresentationareillustratedinapoint-masstask.Second,incomplexhumanoidwalkingexperiments,multiplewalkingpatternswithdifferentstepheightsarelearnedrobustlyandefficiently.Finally,inamulti-directionalreachingtasksimulatedwithamusculoskeletalmodelofthehumanarm,weshowhowtheproposedmovementprimitivescanbeusedtolearnappropriatemuscleexcitationpatternsandtogeneralizeeffectivelytonewreachingskills.

AB - Asalientfeatureofhumanmotorskilllearningistheabilitytoexploitsimilaritiesacrossrelatedtasks.Inbiologicalmotorcontrol,ithasbeenhypothesizedthatmusclesynergies,coherentactivationsofgroupsofmuscles,allowforexploitingsharedknowledge.Recentstudieshaveshownthatarichsetofcomplexmotorskillscanbegeneratedbyacombinationofasmallnumberofmusclesynergies.Inrobotics,dynamicmovementprimitivesarecommonlyusedformotorskilllearning.Thismachinelearningapproachimplementsastableattractorsystemthatfacilitateslearninganditcanbeusedinhigh-dimensionalcontinuousspaces.However,itdoesnotallowforreusingsharedknowledge,i.e.,foreachtaskanindividualsetofparametershastobelearned.Weproposeanovelmovementprimitiverepresentationthatemploysparametrizedbasisfunctions,whichcombinesthebenefitsofmusclesynergiesanddynamicmovementprimitives.Foreachtaskasuperpositionofsynergiesmodulatesastableattractorsystem.Thisapproachleadstoacompactrepresentationofmultiplemotorskillsandatthesametimeenablesefficientlearninginhigh-dimensionalcontinuoussystems.Themovementrepresentationsupportsdiscreteandrhythmicmovementsandinparticularincludesthedynamicmovementprimitiveapproachasaspecialcase.Wedemonstratethefeasibilityofthemovementrepresentationinthreemulti-tasklearningsimulatedscenarios.First,thecharacteristicsoftheproposedrepresentationareillustratedinapoint-masstask.Second,incomplexhumanoidwalkingexperiments,multiplewalkingpatternswithdifferentstepheightsarelearnedrobustlyandefficiently.Finally,inamulti-directionalreachingtasksimulatedwithamusculoskeletalmodelofthehumanarm,weshowhowtheproposedmovementprimitivescanbeusedtolearnappropriatemuscleexcitationpatternsandtogeneralizeeffectivelytonewreachingskills.

UR - https://cps.unileoben.ac.at/wp/Frontiers2013bRueckert.pdf

U2 - 10.3389/fncom.2013.00138

DO - 10.3389/fncom.2013.00138

M3 - Article

VL - 7.2013

JO - Frontiers in computational neuroscience

JF - Frontiers in computational neuroscience

SN - 1662-5188

IS - October

ER -