Tactile guidance for policy refinement and reuse

Brenna D. Argall, Eric L. Sauser, Aude G. Billard

Research output: Chapter in Book/Report/Conference proceedingConference contribution

29 Scopus citations

Abstract

Demonstration learning is a powerful and practical technique to develop robot behaviors. Even so, development remains a challenge and possible demonstration limitations can degrade policy performance. This work presents an approach for policy improvement and adaptation through a tactile interface located on the body of a robot. We introduce the Tactile Policy Correction (TPC) algorithm, that employs tactile feedback for the refinement of a demonstrated policy, as well as its reuse for the development of other policies. We validate TPC on a humanoid robot performing grasp-positioning tasks. The performance of the demonstrated policy is found to improve with tactile corrections. Tactile guidance also is shown to enable the development of policies able to successfully execute novel, undemonstrated, tasks.

Original languageEnglish (US)
Title of host publication2010 IEEE 9th International Conference on Development and Learning, ICDL-2010 - Conference Program
Pages7-12
Number of pages6
DOIs
StatePublished - 2010
Event2010 IEEE 9th International Conference on Development and Learning, ICDL-2010 - Ann Arbor, MI, United States
Duration: Aug 18 2010Aug 21 2010

Publication series

Name2010 IEEE 9th International Conference on Development and Learning, ICDL-2010 - Conference Program

Other

Other2010 IEEE 9th International Conference on Development and Learning, ICDL-2010
Country/TerritoryUnited States
CityAnn Arbor, MI
Period8/18/108/21/10

Keywords

  • Human-robot interaction
  • Humanoid robots
  • Imitation and social learning
  • Skill acquisition

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Software
  • Education

Fingerprint

Dive into the research topics of 'Tactile guidance for policy refinement and reuse'. Together they form a unique fingerprint.

Cite this