Summary
International Symposium on Nonlinear Theory and Its Applications
2015
Session Number:B4L-B
Session:
Number:B4L-B-4
Motion Learning of Robot by Reinforcement Learning Under POMDPs Environments
Ryunosuke Nobori, Yuko Osana,
pp.684-687
Publication Date:2015/12/1
Online ISSN:2188-5079
Summary:
In this paper, motion learning (maze problem) of bipedal walking robot in POMDPs (Partially Observable Markov Decision Processes) environment is realized by the Profit Sharing that can learn deterministic policy for POMDPs environments. In this research, the Profit Sharing that can learn deterministic policy for POMDPs environments which can obtain the deterministic policy by using the history of observations is employed. We carried out a series of experiments using bipedal walking robot, and confirmed that motion learning (maze problem) can be realized by the Profit Sharing that can learn deterministic policy for POMDPs environments.