Value function approximation on non-linear manifolds for robot motor control

M. Sugiyama, H. Hachiya, S. Vijayakumar

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The least squares approach works efficiently in value function approximation, given appropriate basis functions. Because of its smoothness, the Gaussian kernel is a popular and useful choice as a basis function. However, it does not allow for discontinuity which typically arises in real-world reinforcement learning tasks. In this paper, we propose a new basis function based on geodesic Gaussian kernels, which exploits the non-linear manifold structure induced by the Markov decision processes. The usefulness of the proposed method is successfully demonstrated in a simulated robot arm control and Khepera robot navigation.
Original languageEnglish
Title of host publicationRobotics and Automation, 2007 IEEE International Conference on
Pages1733-1740
Number of pages8
ISBN (Electronic)1-4244-0602-1
DOIs
Publication statusPublished - 2007

Fingerprint

Dive into the research topics of 'Value function approximation on non-linear manifolds for robot motor control'. Together they form a unique fingerprint.

Cite this