Full-Body Postural Control of a Humanoid Robot with Both Imitation Learning and Skill Innovation

e-Archivo Repository

Show simple item record

dc.contributor.author González-Fierro Palacios, Miguel
dc.contributor.author Balaguer Bernaldo de Quirós, Carlos
dc.contributor.author Swann, Nicola
dc.contributor.author Nanayakkara, Thrishantha
dc.date.accessioned 2015-02-19T09:03:28Z
dc.date.available 2015-05-01T22:00:06Z
dc.date.issued 2014-04
dc.identifier.bibliographicCitation International Journal of Humanoid Robotics (2014). 11(2).
dc.identifier.issn 0219-8436 (print)
dc.identifier.issn 1793-6942 (online)
dc.identifier.uri http://hdl.handle.net/10016/20064
dc.description.abstract In this paper, we present a novel methodology to obtain imitative and innovative postural movements in a humanoid based on human demonstrations in a different kinematic scale. We collected motion data from a group of human participants standing up from a chair. Modeling the human as an actuated 3-link kinematic chain, and by defining a multi-objective reward function of zero moment point and joint torques to represent the stability and effort, we computed reward profiles for each demonstration. Since individual reward profiles show variability across demonstrating trials, the underlying state transition probabilities were modeled using a Markov chain. Based on the argument that the reward profiles of the robot should show the same temporal structure of those of the human, we used differential evolution to compute a trajectory that fits all humanoid constraints and minimizes the difference between the robot reward profile and the predicted profile if the robot imitates the human. Therefore, robotic imitation involves developing a policy that results in a temporal reward structure, matching that of a group of human demonstrators across an array of demonstrations. Skill innovation was achieved by optimizing a signed reward error after imitation was achieved. Experimental results using the humanoid HOAP-3 are shown.
dc.description.sponsorship The research leading to these results has received funding from the ARCADIA project DPI2010-21047-C02-01, funded by CICYT project grant on behalf of Spanish Ministry of Economy and Competitiveness and from the RoboCity2030-II-CM project (S2009/DPI-1559), funded by Programas de Actividades I+D en la Comunidad de Madrid and co-funded by Structural Funds of the EU.
dc.format.extent 33
dc.format.mimetype application/pdf
dc.language.iso eng
dc.publisher World Scientific Publishing Company
dc.rights © 2014. World Scientific Publishing Company
dc.subject.other Learning from demonstration
dc.subject.other Skill innovation
dc.subject.other Postural control
dc.subject.other Humanoid robot
dc.title Full-Body Postural Control of a Humanoid Robot with Both Imitation Learning and Skill Innovation
dc.type article
dc.description.status Publicado
dc.relation.publisherversion http://dx.doi.org/10.1142/S0219843614500121
dc.subject.eciencia Robótica e Informática Industrial
dc.identifier.doi 10.1142/S0219843614500121
dc.rights.accessRights openAccess
dc.relation.projectID Gobierno de España. DPI2010-21047-C02-01
dc.relation.projectID Comunidad de Madrid. S2009/DPI-1559/ROBOCITY2030 II
dc.type.version acceptedVersion
dc.identifier.publicationfirstpage 1
dc.identifier.publicationissue 2
dc.identifier.publicationlastpage 33
dc.identifier.publicationtitle International Journal of Humanoid Robotics
dc.identifier.publicationvolume 11
dc.identifier.uxxi AR/0000015524
dc.affiliation.dpto UC3M. Departamento de Ingeniería de Sistemas y Automática
dc.affiliation.grupoinv UC3M. Grupo de Investigación: Laboratorio de Robótica (Robotics Lab)
 Find Full text

Files in this item

*Click on file's image for preview. (Embargoed files's preview is not supported)

This item appears in the following Collection(s)

Show simple item record