Header logo is

Reinforcement Learning to adjust Robot Movements to New Situations


Conference Paper


Many complex robot motor skills can be represented using elementary movements, and there exist efficient techniques for learning parametrized motor plans using demonstrations and self-improvement. However with current techniques, in many cases, the robot currently needs to learn a new elementary movement even if a parametrized motor plan exists that covers a related situation. A method is needed that modulates the elementary movement through the meta-parameters of its representation. In this paper, we describe how to learn such mappings from circumstances to meta-parameters using reinforcement learning. In particular we use a kernelized version of the reward-weighted regression. We show two robot applications of the presented setup in robotic domains; the generalization of throwing movements in darts, and of hitting movements in table tennis. We demonstrate that both tasks can be learned successfully using simulated and real robots.

Author(s): Kober, J. and Oztop, E. and Peters, J.
Pages: 2650-2655
Year: 2011
Month: July
Day: 0
Editors: Walsh, T.
Publisher: AAAI Press

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Name: Twenty-Second International Joint Conference on Artificial Intelligence (IJCAI 2011)
Event Place: Barcelona, Spain

Address: Menlo Park, CA, USA
Digital: 0
ISBN: 978-1-57735-512-0

Links: PDF


  title = {Reinforcement Learning to adjust Robot Movements to New Situations},
  author = {Kober, J. and Oztop, E. and Peters, J.},
  pages = {2650-2655},
  editors = {Walsh, T.},
  publisher = {AAAI Press},
  address = {Menlo Park, CA, USA},
  month = jul,
  year = {2011},
  month_numeric = {7}