Automatic design of fuzzy controllers

被引:14
作者
Wang, LX [1 ]
机构
[1] Hong Kong Univ Sci & Technol, Dept Elect & Elect Engn, Clear Water Bay, Kowloon, Peoples R China
关键词
fuzzy control; adaptive control; adaptive algorithms; parameter estimation; nonlinear control;
D O I
10.1016/S0005-1098(99)00044-8
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this note, we propose to use adaptive fuzzy controller as a mean to generate the rule base of fuzzy controller automatically. Specifically, we run an adaptive fuzzy controller from a number of initial conditions such that the trajectories of these runs cover the domain of interest. Since the adaptive fuzzy controller can automatically generate control parameters along the trajectories, we store these parameters after each run. The final fuzzy rule base is constructed from these stored parameters. Since the trajectories cover the whole domain, the rule base generated is complete. The main advantage of the method is that a complete fuzzy rule base can be generated automatically without domain experts and without mathematical model. (C) 1999 Elsevier Science Ltd. All rights reserved.
引用
收藏
页码:1471 / 1475
页数:5
相关论文
共 7 条
[1]  
Astrom K. J., 1995, ADAPTIVE CONTROL
[2]  
Driankov D., 1993, INTRO FUZZY CONTROL
[3]  
Isidori A., 1989, NONLINEAR CONTROL SY
[4]   Conventional fuzzy control and its enhancement [J].
Li, HX ;
Gatland, HB .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 1996, 26 (05) :791-797
[5]  
Wang L.:., 1994, Adaptive Fuzzy System and Control: Desing and Stability Analysis
[6]  
Wang L., 1997, A Course in Fuzzy Systems and Control
[7]   Stable adaptive fuzzy control of nonlinear systems [J].
Wang, Li-Xin .
IEEE Transactions on Fuzzy Systems, 1993, 1 (02) :146-149