New techniques for intelligent control
Chun‐Fei Hsu, Chin‐Teng Lin · 2005
This work proposed a new technique for the intelligent control approach, which is termed as fuzzy-identification-based adaptive fuzzy control (FIAFC) scheme. The developed FIAFC is compared to a principal controller and a robust controller. The principal controller utilizes an adaptive fuzzy model to identify the dynamics of the controlled system. The robust controller is designed to dispel the model error introduced by the adaptive fuzzy model. In the conventional adaptive fuzzy control (AFC) system, the adaptive law was designed to drive the tracking error to zero without considering the modeling error. In the proposed FIAFC system, not only the tracking-error information but also the modeling-error information are utilized in the derived adaptive law, thus the convergence performance can be improved. To investigate the effectiveness of the proposed FIAFC, it is applied to a chaotic system control. The simulation results have demonstrated that the proposed FIAFC can achieve better tracking performance than the AFC. Furthermore, the improvements are achieved at a negligible increase in the computational complexity.