Title :
Optimal design and optimal algorithm of fuzzy CMAC
Author :
Xiaolong, Dai ; Xudong, Zhou ; Meio, A. ; Fukao, T. ; Adachi, N.
Author_Institution :
Kyoto Univ., Japan
Abstract :
A novel CMAC is explored in the paper. The optimal design means to optimize distribution of displacement vectors and membership function of the fuzzy CMAC. This method unifies the optimization problem of fuzzy CMAC and its learning object. The distribution of displacement vectors, the membership function, and optimal internal representation of fuzzy CMAC can be established by genetic algorithm. The learning ability of this optimal fuzzy CMAC is much better than Albus´ CMAC and others. Simulation shows the effectiveness of this method
Keywords :
cerebellar model arithmetic computers; fuzzy neural nets; genetic algorithms; learning (artificial intelligence); displacement vectors; fuzzy CMAC; genetic algorithm; learning ability; membership function; optimal internal representation; optimization problem; Algorithm design and analysis; Design optimization; Extraterrestrial phenomena; Fuzzy systems; Genetic algorithms; Neural networks; Optimization methods; Robot control;
Conference_Titel :
SICE '98. Proceedings of the 37th SICE Annual Conference. International Session Papers
Conference_Location :
Chiba
DOI :
10.1109/SICE.1998.742917
