Learning from Demonstration and Correction via Multiple Modalities for a Humanoid Robot

This paper reports ongoing work that employs multiple demonstration modalities in order to accomplish motion control learning in a multi-staged policy adaptation process. A novel interface for providing tactile guidance to correct learned motion control behaviors is introduced. This interface extend...

Full description

Bibliographic Details
Main Authors: Argall Brenna, Billard Aude
Format: Article
Language:English
Published: EDP Sciences 2011-12-01
Series:BIO Web of Conferences
Online Access:http://dx.doi.org/10.1051/bioconf/20110100003
id doaj-2a49e20e28784e6a91950bb7e06a45bc
record_format Article
spelling doaj-2a49e20e28784e6a91950bb7e06a45bc2021-04-02T18:11:48ZengEDP SciencesBIO Web of Conferences2117-44582011-12-0110000310.1051/bioconf/20110100003Learning from Demonstration and Correction via Multiple Modalities for a Humanoid RobotArgall BrennaBillard AudeThis paper reports ongoing work that employs multiple demonstration modalities in order to accomplish motion control learning in a multi-staged policy adaptation process. A novel interface for providing tactile guidance to correct learned motion control behaviors is introduced. This interface extends our prior work by making use of a more sophisticated set of tactile sensors, developed by the ROBOSKIN consortium.http://dx.doi.org/10.1051/bioconf/20110100003
collection DOAJ
language English
format Article
sources DOAJ
author Argall Brenna
Billard Aude
spellingShingle Argall Brenna
Billard Aude
Learning from Demonstration and Correction via Multiple Modalities for a Humanoid Robot
BIO Web of Conferences
author_facet Argall Brenna
Billard Aude
author_sort Argall Brenna
title Learning from Demonstration and Correction via Multiple Modalities for a Humanoid Robot
title_short Learning from Demonstration and Correction via Multiple Modalities for a Humanoid Robot
title_full Learning from Demonstration and Correction via Multiple Modalities for a Humanoid Robot
title_fullStr Learning from Demonstration and Correction via Multiple Modalities for a Humanoid Robot
title_full_unstemmed Learning from Demonstration and Correction via Multiple Modalities for a Humanoid Robot
title_sort learning from demonstration and correction via multiple modalities for a humanoid robot
publisher EDP Sciences
series BIO Web of Conferences
issn 2117-4458
publishDate 2011-12-01
description This paper reports ongoing work that employs multiple demonstration modalities in order to accomplish motion control learning in a multi-staged policy adaptation process. A novel interface for providing tactile guidance to correct learned motion control behaviors is introduced. This interface extends our prior work by making use of a more sophisticated set of tactile sensors, developed by the ROBOSKIN consortium.
url http://dx.doi.org/10.1051/bioconf/20110100003
work_keys_str_mv AT argallbrenna learningfromdemonstrationandcorrectionviamultiplemodalitiesforahumanoidrobot
AT billardaude learningfromdemonstrationandcorrectionviamultiplemodalitiesforahumanoidrobot
_version_ 1721552241882562560