PID-Like Adaptive Fuzzy Controller Design Based on Absolute Stability Criterion
Jacek Kluska, Tomasz Żabiński · IEEE Transactions on Fuzzy Systems · 2019
This paper describes a method of PID-like adaptive fuzzy controller design for a linear time-invariant single-input and single-output dynamic plant. The plant transfer function is assumed to be partially known. The controller operates in the direct adaptation mode with the output feedback and a nonlinear reference model which provides better closed-loop response parameters than the system with a PID linear controller. The closed-loop stability is guaranteed during the adaptation process. The fuzzy controller is assumed to be a nonlinearity in a bounded sector and is defined by highly interpretable fuzzy rules. The absolute stability criterion for the system, containing PID-like fuzzy controller, is delivered. A design procedure that generalizes and significantly extends results reported so far in the literature and considerably automates the searching process of a nonlinear adaptive fuzzy controller is provided.