Simple record

dc.rights.licenseAttribution-NonCommercial-NoDerivatives 4.0 International*
dc.contributor.authorAbu-Dakka, Fares J.
dc.contributor.otherAnand, Akhil S.
dc.contributor.otherKaushik, Rituraj
dc.contributor.otherGravdahl, Jan Tommy
dc.date.accessioned2024-04-18T12:46:35Z
dc.date.available2024-04-18T12:46:35Z
dc.date.issued2024
dc.identifier.issn2169-3536en
dc.identifier.otherhttps://katalogoa.mondragon.edu/janium-bin/janium_login_opac.pl?find&ficha_no=174317en
dc.identifier.urihttps://hdl.handle.net/20.500.11984/6359
dc.description.abstractOne of the most crucial steps toward achieving human-like manipulation skills in robots is to incorporate compliance into the robot controller. Compliance not only makes the robot’s behaviour safe but also makes it more energy efficient. In this direction, the variable impedance control (VIC) approach provides a framework for a robot to adapt its compliance during execution by employing an adaptive impedance law. Nevertheless, autonomously adapting the compliance profile as demanded by the task remains a challenging problem to be solved in practice. In this work, we introduce a reinforcement learning (RL)-based approach called DEVILC (Data-Efficient Variable Impedance Learning Controller) to learn the variable impedance controller through real-world interaction of the robot. More concretely, we use a model-based RL approach in which, after every interaction, the robot iteratively learns a probabilistic model of its dynamics using the Gaussian process regression model. The model is then used to optimize a neural-network policy that modulates the robot’s impedance such that the long-term reward for the task is maximized. Thanks to the model-based RL framework, DEVILC allows a robot to learn the VIC policy with only a few interactions, making it practical for real-world applications. In simulations and experiments, we evaluate DEVILC on a Franka Emika Panda robotic manipulator for different manipulation tasks in the Cartesian space. The results show that DEVILC is a promising direction toward autonomously learning compliant manipulation skills directly in the real world through interactions. A video of the experiments is available in the link: https://youtu.be/_uyr0Vye5no .en
dc.language.isoengen
dc.publisherIEEEen
dc.rights© 2024 The Authorsen
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectModel-based reinforcement learningen
dc.subjectvariable impedance learning controlen
dc.subjectGaussian processesen
dc.subjectcovariance matrix adaptationen
dc.titleData-efficient reinforcement learning for variable impedance controlen
dcterms.accessRightshttp://purl.org/coar/access_right/c_abf2en
dcterms.sourceIEEE Accessen
local.contributor.groupRobótica y automatizaciónes
local.description.peerreviewedtrueen
local.identifier.doihttps://doi.org/10.1109/ACCESS.2024.3355311en
local.contributor.otherinstitutionhttps://ror.org/020hwjq30en
local.contributor.otherinstitutionhttps://ror.org/05xg72x27en
local.source.detailsVol 12
oaire.format.mimetypeapplication/pdfen
oaire.file$DSPACE\assetstoreen
oaire.resourceTypehttp://purl.org/coar/resource_type/c_6501en
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85en
oaire.funderNameThe Research Council of Norwayen
oaire.funderNameGobierno Vascoen
oaire.funderNameGobierno Vascoen
oaire.funderIdentifierhttps://ror.org/00epmv149
oaire.funderIdentifierhttps://ror.org/00pz2fp31 http://data.crossref.org/fundingdata/funder/10.13039/501100003086
oaire.funderIdentifierhttps://ror.org/00pz2fp31 http://data.crossref.org/fundingdata/funder/10.13039/501100003086
oaire.fundingStreamIKTPLUS-ICT and digital innovationen
oaire.fundingStreamElkartek 2022en
oaire.fundingStreamElkartek 2023en
oaire.awardNumber270941en
oaire.awardNumberKK-2022-00024en
oaire.awardNumberKK-2023-00055en
oaire.awardTitleDynamic Robot Interaction and Motion Compensationen
oaire.awardTitleProducción Fluída y Resiliente para la Industria inteligente (PROFLOW)en
oaire.awardTitleTecnologías de Inteligencia Artificial para la percepción visual y háptica y la planificación y control de tareas de manipulación (HELDU)en
oaire.awardURISin informaciónen
oaire.awardURISin informaciónen
oaire.awardURISin informaciónen


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Simple record

Attribution-NonCommercial-NoDerivatives 4.0 International
Except where otherwise noted, this item's license is described as Attribution-NonCommercial-NoDerivatives 4.0 International