Abstract

Our work aims at developing a robust discriminant controller for robot programming by demonstration. It addresses two core issues of imitation learning, namely "what to imitate" and "how to imitate". This paper presents a method by which a robot extracts the goals of a demonstrated task and determines the imitation strategy that satisfies best these goals. The method is validated in a humanoid platform, taking inspiration of an influential experiment from developmental psychology.

Bibtex reference

@inproceedings{Calinon05,
author = "S. Calinon and F. Guenter and A. Billard",
title = "Goal-Directed Imitation in a Humanoid Robot",
booktitle = "Proceedings of the {IEEE} International Conference on Robotics and Automation ({ICRA})",
year = "2005",
month = "April",
pages = "299--304",
location = "Barcelona, Spain"
}

Video

Goal-directed imitation experiment inspired by studies in developmental psychology from Bekkering et al, by considering a task that consists in reaching dots on a table and by considering different levels of importance for the features to reproduce.