A Sliding Mode Control Learning of Interval Type-2 Intuitionistic Fuzzy Logic for Non-Linear System Prediction


  • Imo Eyoh, Jeremiah Eyoh, Uduak Umoh, Roy Kalawsky


— Many learning algorithms such as gradient descent, extended Kalman filter, decoupled
extended Kalman filter and hybrid approaches have successfully been employed to optimize
intuitionistic fuzzy systems of type-1 and type-2. In this paper, a sliding mode control algorithm for
the optimization of interval type-2 intuitionistic fuzzy logic system parameters is proposed for the
first time. The proposed model adopts the Takagi-Sugeno-type inference. The learning model is
developed and the adaptation for the parameters are derived. The proposed model is applied to nonlinear dynamic prediction problems. Experimental results show that interval type-2 intuitionistic
fuzzy logic system with sliding mode control learning outperforms its type-1 counterpart and some
existing models in the literature while competing favorably with other models on the selected
application domains with less running time.