Self-Scaling Reinforcement Learning Algorithm for Generating Fuzzy Controller

スポンサーリンク

概要

• 論文の詳細を見る

• In this paper, we propose a new reinforcement learning algorithm to generate a fuzzy controller for robot motions. This algorithm generates a range of continuous real-valued actions, and the reinforcement signal is self-scaled. This prevents the weights from overshooting when the system receives very large reinforcement values. Therefore this algorithm can obtain a solution in less iteration times. The proposed method is applied to the control of the brachiation robot, which moves dynamically from branch to branch like a gibbon swinging its body in a pendulum fashion. Through computer simulations, we show the fast convergence and the robustness against disturbances.

• 一般社団法人日本機械学会の論文
• 1997-06-15

著者

• Hasegawa Yasuhisa

  Dept. Of Mechanical Systems Engineering Faculty Of Engineering Gifu University

• Fukuda Toshio

  Dept. Of Mech. Nagoya Univ.

• Fukuda Toshio

  Dept. Micro-Nano Sys. Eng., Nagoya Univ.:Center for Micro-Nano Mech., Nagoya Univ.

関連論文

• 2P313 ドロップレット中への生物時計再構成(非平衡・生体リズム,第48回日本生物物理学会年会)
• Electrostatic Actuator for Moving Cell in Distributed Microrobotic System
• Genetic Evolution and Self-Organization of Cellular Robotic System
• Hierarchical Hybrid Neuromorphic Control System for Robotic Manipulators
• Cell Recognition by Image Processing : Recognition of Dead or Living Plant Cells by Neural Network
• Mobile Mechanism Suitable for Rough Terrain Vehicle
• Application of Image Analysis with Neural Network for Plant Somatic Embryo Culture
• Flexibility Control of Solar Battery Arrays : 2nd Report, Vibration and Attitude Control Based on State Estimation of Differential Solar Cell Sensors
• Learning Method for Multi-Controller of Robot Behavior
• Self-Scaling Reinforcement Learning Algorithm for Generating Fuzzy Controller
• 2A2-2 Strategic Behavior Generation with Cognitive Distance in Two-Player Games
• 1PT229 バクテリア駆動マイクロギアの構築とトルク計測(日本生物物理学会第50回年会(2012年度))

もっと見る 閉じる

スポンサーリンク