Grounded Action Transformation for Robot Learning in Simulation

Josiah Hanna, Peter Stone

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

Abstract / Description of output

Robot learning in simulation is a promising alternative to the prohibitive sample cost of learning in the physical world. Unfortunately, policies learned in simulation often perform worse than hand-coded policies when applied on the physical robot. Grounded simulation learning (GSL) promises to address this issue by altering the simulator to better match the real world. This paper proposes a new algorithm for GSL -- Grounded Action Transformation -- and applies it to learning of humanoid bipedal locomotion. Our approach results in a 43.27% improvement in forward walk velocity compared to a state-of-the art hand-coded walk. We further evaluate our methodology in controlled experiments using a second, higher-fidelity simulator in place of the real world. Our results contribute to a deeper understanding of grounded simulation learning and demonstrate its effectiveness for learning robot control policies.
Original languageEnglish
Title of host publicationProceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17)
PublisherAAAI Press
Pages3834-3840
Number of pages7
Volume5
ISBN (Print)978-1-57735-784-1
Publication statusPublished - 12 Feb 2017
EventThirty-First AAAI Conference on Artificial Intelligence - San Francisco, United States
Duration: 4 Feb 20179 Feb 2017

Publication series

NameProceedings of the AAAI Conference on Artificial Intelligence
PublisherAAAI Press
Volume31
ISSN (Print)2159-5399
ISSN (Electronic)2374-3468

Conference

ConferenceThirty-First AAAI Conference on Artificial Intelligence
Abbreviated titleAAAI 2017
Country/TerritoryUnited States
CitySan Francisco
Period4/02/179/02/17

Keywords / Materials (for Non-textual outputs)

  • Grounded simulation learning
  • Robotic bipedal walking
  • Transfer from simulation

Fingerprint

Dive into the research topics of 'Grounded Action Transformation for Robot Learning in Simulation'. Together they form a unique fingerprint.

Cite this