Presentation 2002/3/11
Motion Control of a Mobile Robot by Reinforcement Learning
Yoshimasa YAMASAKI, Masumi ISHIKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A navigation task for a mobile robot includes path planning to the goal while avoiding obstacles. The present paper aims at training a mobile robot through interation with the envirionment: movement of a mobile robot and reward from the envirionment as a result of movement. This interaction fits well with reinforcement learning. Among various reinforcement learning algorithms Q-learning is adopted here because motion control as well as current position play an important role. A computer simulation demonstrates that Q-learning succeeds in obtaining optimal path to the goal without collision.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) reinforcement learning / mobile robots / Q-learning / optimal path
Paper # NC2001-143
Date of Issue

Conference Information
Committee NC
Conference Date 2002/3/11(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Motion Control of a Mobile Robot by Reinforcement Learning
Sub Title (in English)
Keyword(1) reinforcement learning
Keyword(2) mobile robots
Keyword(3) Q-learning
Keyword(4) optimal path
1st Author's Name Yoshimasa YAMASAKI
1st Author's Affiliation Graduate School of Computer Science and Systems Engineerring Kyushu Institute of Technology()
2nd Author's Name Masumi ISHIKAWA
2nd Author's Affiliation Graduate School of Life Science and Systems Engineerring Kyushu Institute of Technology
Date 2002/3/11
Paper # NC2001-143
Volume (vol) vol.101
Number (no) 735
Page pp.pp.-
#Pages 8
Date of Issue