Design Decoupling Control and Wavelet Type-2 Fuzzy Brain Imitated Neural Network (WT2FBINN) for rotational actuator (TORA) System

Abstract

In this article, a new wavelet type-2 fuzzy brain imitated neural network (WT2FBINN) is presented for nonlinear system decoupling control. A sliding mode control combined with a decoupling control is illustrated. The control system comprises of a proposed WT2FBINN and a robust controller. The WT2FBINN is used as the main controller to approach an ideal controller to achieve desired control performance, and the robust controller is used to eliminate the approximation error to achieve system stability. The WT2FBINN contains a wavelet type-2 fuzzy system and a brain imitated neural network. The parameters of the proposed WT2FBINN can be updated by the gradient descent method. Then, the WIT2FBINN is applied to control underactuated systems, such as a rotational actuator (TORA) system. Finally, simulation results have validated the effectiveness of the proposed control approach. A comparison to a recently developed method is useful for better illustrating the benefits of the proposed method.