Abstract
This paper studies the reinforcement learning (RL) method for central pattern generators (CPG) that generates stable rhythmic movements such as biped locomotion. RL for biped locomotion is very difficult, since the biped robot is highly unstable and the system has continuous state and action spaces with a high degree of freedom. In order to deal with RL for CPG, we propose a new RL method called the CPG-actor-critic method. We applied this method to the RL for the biped robot. The computer simulation showed that our RL method was able to train the CPG such that the biped robot walk stably.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Hirai, K., et al. 1998. Proceedings of ICRA 2:1321–1326
Grillner, S., Wallen, P. and Brodin, L. 1991. Annu. Rev. Neurosci. 14:169–199
Taga, G., Yamaguchi, Y., and Shimizu, H. 1991. Biol. Cybern. 65:147–159
Sutton, R. S. and Barto, A. G. 1998. Reinforcement learning. MIT Press
Sato, M. and Ishii, S. 2000. Neural Computation 12:407–432
Sato, M. and Ishii, S. 1999. NIPS 11, 1052–1058
Morimoto J. and Doya K. 2001. Robot. Auton. Syst., 36:37–51.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sato, Ma., Nakamura, Y., Ishii, S. (2002). Reinforcement Learning for Biped Locomotion. In: Dorronsoro, J.R. (eds) Artificial Neural Networks — ICANN 2002. ICANN 2002. Lecture Notes in Computer Science, vol 2415. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-46084-5_126
Download citation
DOI: https://doi.org/10.1007/3-540-46084-5_126
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-44074-1
Online ISBN: 978-3-540-46084-8
eBook Packages: Springer Book Archive