Publication

Active learning of manipulation sequences

Conference Article

Conference

IEEE International Conference on Robotics and Automation (ICRA)

Edition

2014

Pages

5671-5678

Doc link

http://dx.doi.org/10.1109/ICRA.2014.6907693

File

Download the digital copy of the doc pdf document

Authors

Projects associated

Abstract

We describe a system allowing a robot to learn goal-directed manipulation sequences such as steps of an assembly task. Learning is based on a free mix of exploration and instruction by an external teacher, and may be active in the sense that the system tests actions to maximize learning progress and asks the teacher if needed. The main component is a symbolic planning engine that operates on learned rules, defined by actions and their pre- and postconditions. Learned by model-based reinforcement learning, rules are immediately available for planning. Thus, there are no distinct learning and application phases. We show how dynamic plans, replanned after every action if necessary, can be used for automatic execution of manipulation sequences, for monitoring of observed manipulation sequences, or a mix of the two, all while extending and refining the rule base on the fly. Quantitative results indicate fast convergence using few training examples, and highly effective teacher intervention at early stages of learning.

Categories

learning (artificial intelligence), planning (artificial intelligence), uncertainty handling.

Scientific reference

D. Martínez, G. Alenyà, P. Jiménez, C. Torras, J. Rossmann, N. Wantia, A. Eren Erdal, S. Haller and J. Piater. Active learning of manipulation sequences, 2014 IEEE International Conference on Robotics and Automation, 2014, Hong Kong, pp. 5671-5678.