A study on multi-dimensional fuzzy Q-learning for intelligent robots

Kazuo Kiguchi, He Hui, Kenbu Teramoto

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Reinforcement learning is one of the most important learning methods for intelligent robots working in unknown/uncertain environments. Multi-dimensional fuzzy Q-learning, an extension of the Q-learning method, has been proposed in this study. The proposed method has been applied for an intelligent robot working in a dynamic environment. The rewards from the evaluation functions and the fuzzy Q-values generated by the neural networks (fuzzy Q-net) are expressed in vector forms in order to obtain optimal behaviors for several different purposes. By applying this learning method, evaluation and learning of fuzzy Q-values for the other behaviors can be carried out simultaneously in one trial. We express fuzzy states as the vector of fuzzy sets for input variables of the fuzzy Q-net. The behavior selection algorithm is also proposed in this study. The simulation results show the effectives of the proposed methods for a mobile robot selects optimal behavior in a dynamic environment.

Original languageEnglish
Pages (from-to)95-104
Number of pages10
JournalInternational Journal of Fuzzy Systems
Volume9
Issue number2
Publication statusPublished - Jun 1 2007

Fingerprint

Intelligent robots
Q-learning
Robot
Function evaluation
Fuzzy neural networks
Reinforcement learning
Fuzzy sets
Mobile robots
Dynamic Environment
Fuzzy Neural Network
Evaluation Function
Evaluation Method
Reinforcement Learning
Reward
Mobile Robot
Fuzzy Sets
Express
Unknown
Learning
Simulation

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Computational Theory and Mathematics
  • Artificial Intelligence

Cite this

A study on multi-dimensional fuzzy Q-learning for intelligent robots. / Kiguchi, Kazuo; Hui, He; Teramoto, Kenbu.

In: International Journal of Fuzzy Systems, Vol. 9, No. 2, 01.06.2007, p. 95-104.

Research output: Contribution to journalArticle

@article{0193b0389d7640f69f9b2349d310e61c,
title = "A study on multi-dimensional fuzzy Q-learning for intelligent robots",
abstract = "Reinforcement learning is one of the most important learning methods for intelligent robots working in unknown/uncertain environments. Multi-dimensional fuzzy Q-learning, an extension of the Q-learning method, has been proposed in this study. The proposed method has been applied for an intelligent robot working in a dynamic environment. The rewards from the evaluation functions and the fuzzy Q-values generated by the neural networks (fuzzy Q-net) are expressed in vector forms in order to obtain optimal behaviors for several different purposes. By applying this learning method, evaluation and learning of fuzzy Q-values for the other behaviors can be carried out simultaneously in one trial. We express fuzzy states as the vector of fuzzy sets for input variables of the fuzzy Q-net. The behavior selection algorithm is also proposed in this study. The simulation results show the effectives of the proposed methods for a mobile robot selects optimal behavior in a dynamic environment.",
author = "Kazuo Kiguchi and He Hui and Kenbu Teramoto",
year = "2007",
month = "6",
day = "1",
language = "English",
volume = "9",
pages = "95--104",
journal = "International Journal of Fuzzy Systems",
issn = "1562-2479",
publisher = "Chinese Fuzzy Systems Association",
number = "2",

}

TY - JOUR

T1 - A study on multi-dimensional fuzzy Q-learning for intelligent robots

AU - Kiguchi, Kazuo

AU - Hui, He

AU - Teramoto, Kenbu

PY - 2007/6/1

Y1 - 2007/6/1

N2 - Reinforcement learning is one of the most important learning methods for intelligent robots working in unknown/uncertain environments. Multi-dimensional fuzzy Q-learning, an extension of the Q-learning method, has been proposed in this study. The proposed method has been applied for an intelligent robot working in a dynamic environment. The rewards from the evaluation functions and the fuzzy Q-values generated by the neural networks (fuzzy Q-net) are expressed in vector forms in order to obtain optimal behaviors for several different purposes. By applying this learning method, evaluation and learning of fuzzy Q-values for the other behaviors can be carried out simultaneously in one trial. We express fuzzy states as the vector of fuzzy sets for input variables of the fuzzy Q-net. The behavior selection algorithm is also proposed in this study. The simulation results show the effectives of the proposed methods for a mobile robot selects optimal behavior in a dynamic environment.

AB - Reinforcement learning is one of the most important learning methods for intelligent robots working in unknown/uncertain environments. Multi-dimensional fuzzy Q-learning, an extension of the Q-learning method, has been proposed in this study. The proposed method has been applied for an intelligent robot working in a dynamic environment. The rewards from the evaluation functions and the fuzzy Q-values generated by the neural networks (fuzzy Q-net) are expressed in vector forms in order to obtain optimal behaviors for several different purposes. By applying this learning method, evaluation and learning of fuzzy Q-values for the other behaviors can be carried out simultaneously in one trial. We express fuzzy states as the vector of fuzzy sets for input variables of the fuzzy Q-net. The behavior selection algorithm is also proposed in this study. The simulation results show the effectives of the proposed methods for a mobile robot selects optimal behavior in a dynamic environment.

UR - http://www.scopus.com/inward/record.url?scp=40849134332&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=40849134332&partnerID=8YFLogxK

M3 - Article

VL - 9

SP - 95

EP - 104

JO - International Journal of Fuzzy Systems

JF - International Journal of Fuzzy Systems

SN - 1562-2479

IS - 2

ER -