Recursive filtering adaptive neural fault-tolerant control for uncertain multivariable nonlinear systems

Abstract

This paper synthesizes a recursive filtering adaptive neural fault-tolerant controller for uncertain multivariable nonlinear systems. The proposed control scheme adopts an estimation/cancellation strategy to deal with uncertainties and/or disturbances. The nonlinear uncertainties are approximated by a Gaussian radial basis function (GRBF)-based neural network incorporated with a piecewise constant adaptive law, where the adaptive law will generate adaptive parameters by solving the error dynamics between the real system and the state predictor with the neglection of unknowns, and recursive least squares (RLS) is applied to distribute the total uncertainty estimates into matched and mismatched components. The cooperation of GRBF learning method and piecewise constant adaptive law relaxes the stringent constraint on the hardware CPU speed and achieves fast adaptation. The filtering control law delivers a satisfactory performance with guaranteed robustness. Two numerical examples are provided to illustrate the effectiveness of the proposed control architecture via comparisons.