Self-Evolving Interval Type-2 Wavelet Cerebellar Model Articulation Control Design for Uncertain Nonlinear Systems Using PSO

Abstract

This paper aims to propose a design method of interval type-2 wavelet cerebellar model articulation controller (IT2WCMAC) and applies it to control uncertain nonlinear systems. The proposed controller incorporates an IT2WCMAC as the main controller to mimic an ideal controller, and a robust compensator is used to eliminate the approximation error between the IT2WCMAC and the ideal controller. The self-evolving algorithm is used to automatically construct the network structure from the blank rule-base. In the proposed control scheme, the steepest descent gradient algorithm is applied to online tune the network parameters, and a Lyapunov stability theorem is applied to guarantee the system’s stability. Moreover, the learning rates of the parameter adaptive laws can be optimized by the particle swarm optimization (PSO) to promote the parameter learning efficiency. Finally, the proposed control system is applied to control nonlinear chaotic systems to verify the control performance and to show the superiority of the proposed algorithm than the other control methods.