Learning from Demonstration: Communication and Policy Generation

Abstract

Abstract Learning from demonstration utilizes human expertise to program a robot. We believe this approach to robot programming will facilitate the development and deployment of general purpose personal robots that can adapt to specific user preferences. Demonstrations can potentially take place across a wide variety of environmental conditions. In this paper we address how learning from demonstration can be affected by various communication alterations. Furthermore, we we detail a Bayesian approach to generating task policies from demonstration data. 1