A formal approach to fuzzy modeling

被引:16
作者
Lygeros, J
机构
[1] Department of Electrical Engineering and Computer Science, Laboratory for Computer Science, Massachusetts Institute of Technology, Cambridge
关键词
dynamical systems; function approximation; fuzzy modeling;
D O I
10.1109/91.618270
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A formalism for coding fuzzy models of dynamical systems is presented. It is shown that the formalism is rich enough to capture the performance of arbitrary conventional discrete time dynamical systems whose transition maps are polynomials with rational coefficients. The proof of this fact provides a constructive algorithm for generating fuzzy models to arbitrarily closely approximate an arbitrary map on a compact set. Our modeling formalism highlights the similarities between fuzzy systems and hybrid control systems, We hope to be able to exploit these similarities by extending results from the area of hybrid systems to the fuzzy domain and vice versa.
引用
收藏
页码:317 / 327
页数:11
相关论文
共 17 条
[1]  
BROCKETT RW, 1993, PERSPECTIVES CONTROL
[2]  
Callier F. M., 1991, Linear System Theory
[3]  
Driankov D., 1993, INTRO FUZZY CONTROL
[4]  
Grossman RobertL., 1993, HYBRID SYSTEMS, V736
[5]   FUZZY-SYSTEMS AS UNIVERSAL APPROXIMATORS [J].
KOSKO, B .
IEEE TRANSACTIONS ON COMPUTERS, 1994, 43 (11) :1329-1333
[6]  
KRSTIC M, 1995, NONLINEAR ADAPTATIVE
[7]  
LYGEROS J, 1994, UCBERLM9460
[8]   EXPERIMENT IN LINGUISTIC SYNTHESIS WITH A FUZZY LOGIC CONTROLLER [J].
MAMDANI, EH ;
ASSILIAN, S .
INTERNATIONAL JOURNAL OF MAN-MACHINE STUDIES, 1975, 7 (01) :1-13
[9]  
Nerode A., 1993, Hybrid Systems, P317
[10]  
PASSINO KM, 1993, INTRO INTELLIGENT AU, P191