Show simple item record

dc.contributor.authorMartínez Martínez, David
dc.contributor.authorAlenyà Ribas, Guillem
dc.contributor.authorJimenez Schlegl, Pablo
dc.contributor.authorTorras, Carme
dc.contributor.authorRossmann, Jürgen
dc.contributor.authorWantia, Nils
dc.contributor.authorEren Erdal, Aksoy
dc.contributor.authorHaller, Simon
dc.contributor.authorPiater, Justus
dc.contributor.otherInstitut de Robòtica i Informàtica Industrial
dc.date.accessioned2015-09-03T14:21:07Z
dc.date.available2015-09-03T14:21:07Z
dc.date.issued2014
dc.identifier.citationMartínez, D., Alenyà, G., Jimenez, P., Torras, C., Rossmann, J., Wantia, N., Eren Erdal, A., Haller, S., Piater, J. Active learning of manipulation sequences. A: IEEE International Conference on Robotics and Automation. "Proceedings of the ICRA - 2014 - IEEE International Conference on Robotics and Automation". Hong Kong: 2014, p. 5671-5678.
dc.identifier.urihttp://hdl.handle.net/2117/76605
dc.description.abstractWe describe a system allowing a robot to learn goal-directed manipulation sequences such as steps of an assembly task. Learning is based on a free mix of exploration and instruction by an external teacher, and may be active in the sense that the system tests actions to maximize learning progress and asks the teacher if needed. The main component is a symbolic planning engine that operates on learned rules, defined by actions and their pre- and postconditions. Learned by model-based reinforcement learning, rules are immediately available for planning. Thus, there are no distinct learning and application phases. We show how dynamic plans, replanned after every action if necessary, can be used for automatic execution of manipulation sequences, for monitoring of observed manipulation sequences, or a mix of the two, all while extending and refining the rule base on the fly. Quantitative results indicate fast convergence using few training examples, and highly effective teacher intervention at early stages of learning.
dc.format.extent8 p.
dc.language.isoeng
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/3.0/es/
dc.subjectÀrees temàtiques de la UPC::Informàtica::Intel·ligència artificial
dc.subject.otherlearning (artificial intelligence)
dc.subject.otherplanning (artificial intelligence)
dc.subject.otheruncertainty handling.
dc.titleActive learning of manipulation sequences
dc.typeConference report
dc.contributor.groupUniversitat Politècnica de Catalunya. ROBiri - Grup de Robòtica de l'IRI
dc.identifier.doi10.1109/ICRA.2014.6907693
dc.description.peerreviewedPeer Reviewed
dc.subject.inspecClassificació INSPEC::Cybernetics::Artificial intelligence
dc.relation.publisherversionhttp://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6907693
dc.rights.accessOpen Access
local.identifier.drac15271713
dc.description.versionPostprint (author’s final draft)
dc.relation.projectidinfo:eu-repo/grantAgreement/EC/FP7/269959/EU/Intelligent observation and execution of Actions and manipulations/INTELLACT
local.citation.authorMartínez, D.; Alenyà, G.; Jimenez, P.; Torras, C.; Rossmann, J.; Wantia, N.; Eren Erdal, A.; Haller, S.; Piater, J.
local.citation.contributorIEEE International Conference on Robotics and Automation
local.citation.pubplaceHong Kong
local.citation.publicationNameProceedings of the ICRA - 2014 - IEEE International Conference on Robotics and Automation
local.citation.startingPage5671
local.citation.endingPage5678


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

Attribution-NonCommercial-NoDerivs 3.0 Spain
Except where otherwise noted, content on this work is licensed under a Creative Commons license : Attribution-NonCommercial-NoDerivs 3.0 Spain