Controlled Gaussian process dynamical models with application to robotic cloth manipulation

Over the last years, significant advances have been made in robotic manipulation, but still, the handling of non-rigid objects, such as cloth garments, is an open problem. Physical interaction with non-rigid objects is uncertain and complex to model. Thus, extracting useful information from sample d...

ver descrição completa

Detalhes bibliográficos
Autores: Amadio, Fabio, Delgado-Guerrero, Juan Antonio, Colomé, Adrià, Torras, Carme
Tipo de documento: artigo
Estado:Versão publicada
Data de publicação:2023
País:España
Recursos:Consejo Superior de Investigaciones Científicas (CSIC)
Repositório:DIGITAL.CSIC. Repositorio Institucional del CSIC
OAI Identifier:oai:digital.csic.es:10261/351159
Acesso em linha:http://hdl.handle.net/10261/351159
https://api.elsevier.com/content/abstract/scopus_id/85159334069
Access Level:Acceso aberto
Palavra-chave:Data-driven modeling
Dimensionality reduction
Gaussian processes
High-dimensional dynamical systems
id ES_a3c7da8b6a1c38b58189fe59c05a3769
oai_identifier_str oai:digital.csic.es:10261/351159
network_acronym_str ES
network_name_str España
repository_id_str
spelling Controlled Gaussian process dynamical models with application to robotic cloth manipulationAmadio, FabioDelgado-Guerrero, Juan AntonioColomé, AdriàTorras, CarmeData-driven modelingDimensionality reductionGaussian processesHigh-dimensional dynamical systemsOver the last years, significant advances have been made in robotic manipulation, but still, the handling of non-rigid objects, such as cloth garments, is an open problem. Physical interaction with non-rigid objects is uncertain and complex to model. Thus, extracting useful information from sample data can considerably improve modeling performance. However, the training of such models is a challenging task due to the high-dimensionality of the state representation. In this paper, we propose Controlled Gaussian Process Dynamical Models (CGPDMs) for learning high-dimensional, nonlinear dynamics by embedding them in a low-dimensional manifold. A CGPDM is constituted by a low-dimensional latent space, with an associated dynamics where external control variables can act and a mapping to the observation space. The parameters of both maps are marginalized out by considering Gaussian Process priors. Hence, a CGPDM projects a high-dimensional state space into a smaller dimension latent space, in which it is feasible to learn the system dynamics from training data. The modeling capacity of CGPDM has been tested in both a simulated and a real scenario, where it proved to be capable of generalizing over a wide range of movements and confidently predicting the cloth motions obtained by previously unseen sequences of control actions.Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. This work was partially developed in the context of the project CLOTHILDE (“CLOTH manIpulation Learning from DEmonstrations”), which has received funding from ERC under the European Union’s Horizon 2020 research and innovation program (Advanced Grant agreement No 741930). This work has also received funding from project CHLOE-GRAPH (PID2020-118649RB-I00) funded by MCIN/ AEI /10.13039/501100011033.Peer reviewedSpringerConferencia de Rectores de las Universidades EspañolasConsejo Superior de Investigaciones Científicas (España)European Research CouncilEuropean CommissionAgencia Estatal de Investigación (España)Ministerio de Ciencia, Innovación y Universidades (España)Amadio, Fabio [0000-0002-0866-2952]Delgado-Guerrero, Juan Antonio [0000-0001-6682-8810]Colomé, Adrià [0000-0001-9715-4062]Torras, Carme [0000-0002-2933-398X]Consejo Superior de Investigaciones Científicas [https://ror.org/02gfc7t72]202420242023info:eu-repo/semantics/articlehttp://purl.org/coar/resource_type/c_6501Publisher's versioninfo:eu-repo/semantics/publishedVersionapplication/pdfhttp://hdl.handle.net/10261/351159https://api.elsevier.com/content/abstract/scopus_id/85159334069reponame:DIGITAL.CSIC. Repositorio Institucional del CSICinstname:Consejo Superior de Investigaciones Científicas (CSIC)Inglés#PLACEHOLDER_PARENT_METADATA_VALUE##PLACEHOLDER_PARENT_METADATA_VALUE#info:eu-repo/grantAgreement/EC/H2020/741930info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020/PID2020-118649RB-I00The underlying dataset has been published as supplementary material of the article in the publisher platform at https://doi.org/10.1007/s40435-023-01205-6https://doi.org/10.1007/s40435-023-01205-6Síinfo:eu-repo/semantics/openAccessoai:digital.csic.es:10261/3511592026-05-22T06:33:51Z
dc.title.none.fl_str_mv Controlled Gaussian process dynamical models with application to robotic cloth manipulation
title Controlled Gaussian process dynamical models with application to robotic cloth manipulation
spellingShingle Controlled Gaussian process dynamical models with application to robotic cloth manipulation
Amadio, Fabio
Data-driven modeling
Dimensionality reduction
Gaussian processes
High-dimensional dynamical systems
title_short Controlled Gaussian process dynamical models with application to robotic cloth manipulation
title_full Controlled Gaussian process dynamical models with application to robotic cloth manipulation
title_fullStr Controlled Gaussian process dynamical models with application to robotic cloth manipulation
title_full_unstemmed Controlled Gaussian process dynamical models with application to robotic cloth manipulation
title_sort Controlled Gaussian process dynamical models with application to robotic cloth manipulation
dc.creator.none.fl_str_mv Amadio, Fabio
Delgado-Guerrero, Juan Antonio
Colomé, Adrià
Torras, Carme
author Amadio, Fabio
author_facet Amadio, Fabio
Delgado-Guerrero, Juan Antonio
Colomé, Adrià
Torras, Carme
author_role author
author2 Delgado-Guerrero, Juan Antonio
Colomé, Adrià
Torras, Carme
author2_role author
author
author
dc.contributor.none.fl_str_mv Conferencia de Rectores de las Universidades Españolas
Consejo Superior de Investigaciones Científicas (España)
European Research Council
European Commission
Agencia Estatal de Investigación (España)
Ministerio de Ciencia, Innovación y Universidades (España)
Amadio, Fabio [0000-0002-0866-2952]
Delgado-Guerrero, Juan Antonio [0000-0001-6682-8810]
Colomé, Adrià [0000-0001-9715-4062]
Torras, Carme [0000-0002-2933-398X]
Consejo Superior de Investigaciones Científicas [https://ror.org/02gfc7t72]
dc.subject.none.fl_str_mv Data-driven modeling
Dimensionality reduction
Gaussian processes
High-dimensional dynamical systems
topic Data-driven modeling
Dimensionality reduction
Gaussian processes
High-dimensional dynamical systems
description Over the last years, significant advances have been made in robotic manipulation, but still, the handling of non-rigid objects, such as cloth garments, is an open problem. Physical interaction with non-rigid objects is uncertain and complex to model. Thus, extracting useful information from sample data can considerably improve modeling performance. However, the training of such models is a challenging task due to the high-dimensionality of the state representation. In this paper, we propose Controlled Gaussian Process Dynamical Models (CGPDMs) for learning high-dimensional, nonlinear dynamics by embedding them in a low-dimensional manifold. A CGPDM is constituted by a low-dimensional latent space, with an associated dynamics where external control variables can act and a mapping to the observation space. The parameters of both maps are marginalized out by considering Gaussian Process priors. Hence, a CGPDM projects a high-dimensional state space into a smaller dimension latent space, in which it is feasible to learn the system dynamics from training data. The modeling capacity of CGPDM has been tested in both a simulated and a real scenario, where it proved to be capable of generalizing over a wide range of movements and confidently predicting the cloth motions obtained by previously unseen sequences of control actions.
publishDate 2023
dc.date.none.fl_str_mv 2023
2024
2024
dc.type.none.fl_str_mv info:eu-repo/semantics/article
http://purl.org/coar/resource_type/c_6501
Publisher's version
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10261/351159
https://api.elsevier.com/content/abstract/scopus_id/85159334069
url http://hdl.handle.net/10261/351159
https://api.elsevier.com/content/abstract/scopus_id/85159334069
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv #PLACEHOLDER_PARENT_METADATA_VALUE#
#PLACEHOLDER_PARENT_METADATA_VALUE#
info:eu-repo/grantAgreement/EC/H2020/741930
info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020/PID2020-118649RB-I00
The underlying dataset has been published as supplementary material of the article in the publisher platform at https://doi.org/10.1007/s40435-023-01205-6
https://doi.org/10.1007/s40435-023-01205-6
Sí
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Springer
publisher.none.fl_str_mv Springer
dc.source.none.fl_str_mv reponame:DIGITAL.CSIC. Repositorio Institucional del CSIC
instname:Consejo Superior de Investigaciones Científicas (CSIC)
instname_str Consejo Superior de Investigaciones Científicas (CSIC)
reponame_str DIGITAL.CSIC. Repositorio Institucional del CSIC
collection DIGITAL.CSIC. Repositorio Institucional del CSIC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869415425675100160
score 15.198674

Registros relacionados