Adaptive Neuro–Fuzzy Control for Discrete-Time Nonaffine Nonlinear Systems

Authors

Abstract

The problem of nonaffine time-varying nonlinear control systems is addressed in this paper through an adaptive state-space neuro-fuzzy control scheme. It combines an eight-layered neuro-fuzzy model to approximate nonaffine nonlinear systems' dynamics with a state feedback quadratic stabilizing controller. Both the neuro-fuzzy model and controller are updated online within a constrained unscented Kalman filter framework. The proposed generalized state-space neuro-fuzzy model is shown to be an universal approximator, and stability conditions derived for time-varying closed-loop systems. Results from a benchmark multi-input and multi-output system demonstrate the effectiveness of the proposed approach.

Keywords

Adaptation models, adaptive control, adaptive state-space neuro-fuzzy control scheme, Adaptive systems, approximate nonaffine nonlinear systems, approximation theory, closed loop systems, discrete time systems, discrete-time nonaffine nonlinear systems, fuzzy control, fuzzy neural nets, Fuzzy systems, Kalman filter, Kalman filters, MIMO systems, multiinput-and-multioutput system, neuro–fuzzy control, neurocontrollers, Noise measurement, nonlinear control systems, nonlinear filters, Nonlinear systems, nonlinear time-varying systems, recursive learning, stability, Stability analysis, stability conditions, state feedback, state feedback quadratic stabilizing controller, state-space methods, time-varying closed-loop systems, time-varying systems, Time-varying systems, unscented Kalman filter, unscented transform

Journal

IEEE Transactions on Fuzzy Systems, Vol. 27, #8, pp. 1602-1615 2019

DOI