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.
[1]
Li-Xin Wang.
Stable adaptive fuzzy control of nonlinear systems
,
1993,
IEEE Trans. Fuzzy Syst..
[2]
Karl Johan Åström,et al.
Adaptive Control
,
1989,
Embedded Digital Control with Microcontrollers.
[3]
Han-Xiong Li,et al.
Conventional fuzzy control and its enhancement
,
1996,
IEEE Trans. Syst. Man Cybern. Part B.
[4]
Dr. Hans Hellendoorn,et al.
An Introduction to Fuzzy Control
,
1996,
Springer Berlin Heidelberg.
[5]
Alberto Isidori,et al.
Nonlinear control systems: an introduction (2nd ed.)
,
1989
.