IEEE ICRA`07 Digest Template

Towards a Real-Time Bayesian Imitation System
for a Humanoid Robot
Aaron P. Shon, Joshua J. Storz, Rajesh P.N. Rao
Department of Computer Science and Engineering, University of Washington,
Seattle WA USA
• Imitation learning, or programming by
demonstration, could allow robots to acquire
domain-specific skills directly from humans.
• We have prototyped a real-time closed-loop
system for teaching a humanoid robot to
interact with objects in its environment.
• Nonparametric Bayesian inference is used to
determine an optimal action (out of directly
observed actions) given current object
configuration and a goal.