Presentation 2002/7/19
Navigation of Mobile Robot using Neural-Gas and Reinforcement Learning
Toshio TANAKA, Kenji NISHIDA, Takio KURITA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Martinetz et al. proposed a neural network model which can preserve the topology of the place cells and can be self-organized by using neural-gas and competitive Hebbian learning. In his computer simulation on exploration of a rat in an environment consisted of the square area with a number of obstacles, it is shown that both a topology preserving map and a path preserving representation of the obstacle-free area can be self-organized by the network. In this paper, we propose a learning algorithm for navigation on the Martinetz's network based on the actor-critic reinforcement learning. Our simulation results shows that the time to arrive at the goal gradually improved as the learning proceeds.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) neural-gas / place cell / mobile robot / navigation / reinforcement learning
Paper # NC2002-41
Date of Issue

Conference Information
Committee NC
Conference Date 2002/7/19(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) Navigation of Mobile Robot using Neural-Gas and Reinforcement Learning
Sub Title (in English)
Keyword(1) neural-gas
Keyword(2) place cell
Keyword(3) mobile robot
Keyword(4) navigation
Keyword(5) reinforcement learning
1st Author's Name Toshio TANAKA
1st Author's Affiliation National Institute of Advanced Industrial Science and Technology()
2nd Author's Name Kenji NISHIDA
2nd Author's Affiliation National Institute of Advanced Industrial Science and Technology
3rd Author's Name Takio KURITA
3rd Author's Affiliation National Institute of Advanced Industrial Science and Technology
Date 2002/7/19
Paper # NC2002-41
Volume (vol) vol.102
Number (no) 253
Page pp.pp.-
#Pages 6
Date of Issue