March
2015
Efficient Model Learning from Joint-Action Demonstrations for Human-Robot Collaborative Tasks
Authors:
Abstract:
We present a framework for automatically learning human user models from joint-action demonstrations that enables a robot to compute a robust policy for a collaborative task with a human. First, the demonstrated action sequences are clustered into different human types using an unsupervised learning algorithm. A reward function is then learned for each type through the employment of an inverse reinforcement learning algorithm. The learned model is then incorporated into a mixed-observability Markov decision process (MOMDP) formulation, wherein the human type is a partially observable variable. With this framework, we can infer online the human type of a new user that was not included in the training set, and can compute a policy for the robot that will be aligned to the preference of this user. In a human subject experiment (n=30), participants agreed more strongly that the robot anticipated their actions when working with a robot incorporating the proposed framework (p
Notes:
Best Enabling Technology Award
copied = false, 2000);
">
@conference{Nikolaidis-2015-5931,
author = {Stefanos Nikolaidis And Ramya Ramakrishnan And Keren Gu And Julie Shah},
title = {Efficient Model Learning from Joint-Action Demonstrations for Human-Robot Collaborative Tasks},
booktitle = {Proceedings of 10th Annual ACM/IEEE International Conference on Human-Robot Interaction (HRI '15)},
year = {2015},
month = {March},
pages = {189 - 196},
}
author = {Stefanos Nikolaidis And Ramya Ramakrishnan And Keren Gu And Julie Shah},
title = {Efficient Model Learning from Joint-Action Demonstrations for Human-Robot Collaborative Tasks},
booktitle = {Proceedings of 10th Annual ACM/IEEE International Conference on Human-Robot Interaction (HRI '15)},
year = {2015},
month = {March},
pages = {189 - 196},
}