Header logo is

Discovering optimal imitation strategies




This paper develops a general policy for learning relevant features of an imitation task. We restrict our study to imitation of manipulative tasks or of gestures. The imitation process is modeled as a hierarchical optimization system, which minimizes the discrepancy between two multi-dimensional datasets. To classify across manipulation strategies, we apply a probabilistic analysis to data in Cartesian and joint spaces. We determine a general metric that optimizes the policy of task reproduction, following strategy determination. The model successfully discovers strategies in six different imitative tasks and controls task reproduction by a full body humanoid robot.

Author(s): Billard, A. and Epars, Y. and Calinon, S. and Cheng, G. and Schaal, S.
Book Title: Robotics and Autonomous Systems
Volume: 47
Number (issue): 2-3
Pages: 68-77
Year: 2004

Department(s): Autonomous Motion
Bibtex Type: Article (article)

Cross Ref: p1959
Note: clmc


  title = {Discovering optimal imitation strategies},
  author = {Billard, A. and Epars, Y. and Calinon, S. and Cheng, G. and Schaal, S.},
  booktitle = {Robotics and Autonomous Systems},
  volume = {47},
  number = {2-3},
  pages = {68-77},
  year = {2004},
  note = {clmc},
  crossref = {p1959}