Adaptive state construction for reinforcement learning and its application to robot navigation problems

Proceedings of the IEEE International Conference on Systems, Man and Cybernetics

Volumn 3, Issue , 2001, Pages 1436-1441

  Handa, Hisashi ^a   Ninomiya, Akira ^a   Horiuchi, Tadashi ^b   Konishi, Tadataka ^c   Baba, Mitsuru ^a  

^a OKAYAMA UNIVERSITY   (Japan)

^b MATSUE NATIONAL COLLEGE OF TECHNOLOGY   (Japan)

^c Chugoku Polytechnic College   (Brazil)

Author keywords

Adaptive state construction; ART neural network; Reinforcement learning

Indexed keywords

ADAPTIVE CONTROL SYSTEMS; COMPUTER SIMULATION; LEARNING ALGORITHMS; NEURAL NETWORKS; ROBOT APPLICATIONS;

REINFORCEMENT LEARNING; ROBOT NAVIGATION;

LEARNING SYSTEMS;

EID: 0035720226     PISSN: 08843627     EISSN: None     Source Type: Journal    
DOI: 10.1109/ICSMC.2001.973484     Document Type: Article

References (12)

1. An incremental state-segmantation method for reinforcement learning using ART neural network
2. Q-learning
3. Learning to predict by the method of temporal differences
4. Learning to categorize perceptual space of a mobile robot using fuzzy-ART neural network
5. Incremental state acquisition for Q-learning by adaptive gaussian soft-max neural network
6. Incremental quantization of the continuous sensor space for leaning agents
7. Reinforcement learning with autonomous state space construction using unsupervised clustering method
8. Reinforcement learning with perceptual aliasing
9. Generalized ART2 algorithms