Recursive filtering adaptive fault‐tolerant tracking control for uncertain switched multivariable nonlinear systems

Abstract

This article synthesizes a recursive filtering adaptive fault-tolerant tracking control method for uncertain switched multivariable nonlinear systems. The multivariable nonlinear systems under consideration have both matched and mismatched uncertainties, which satisfy the semiglobal Lipschitz condition. A piecewise constant adaptive law generates adaptive parameters by solving the error dynamics with the neglection of unknowns, and the recursive least squares is employed to minimize the residual error by categorizing the total uncertainty estimates into matched and mismatched components. A filtering control law is designed to compensate the actuator faults and nonlinear uncertainties such that a good tracking performance is delivered with guaranteed robustness. The matched component is canceled directly by adopting their opposite in the control signal, whereas a dynamic inversion of the system is performed to eliminate the effect of the mismatched component on the output. By exploiting the average dwell time principle, the error bounds are derived for the states and control inputs compared with the virtual reference system which defines the best performance that can be achieved by the closed-loop system. Both numerical and practical examples are provided to illustrate the effectiveness of the proposed switching recursive filtering adaptive fault-tolerant tracking control architecture, comparisons with model reference adaptive control are also carried out.

Adaptive fixed-time control for nonlinear systems against time-varying actuator faults

“…Therefore, it is significant to improve system reliability and safety not only by designing fault tolerant control (FTC) to compensate the effect caused by faults automatically, but also by enhancing reliability of signal components. The approaches of the fault-tolerant design can be broadly divided into two categories: the passive one [1][2][3][4][5] and the active one [6][7][8][9][10][11]. Although the passive approach is usually exploited to handle partial actuator faults [12][13][14] and complete actuator faults [15][16][17], it has also a limited capability of handling unknown actuator faults due to its passive control laws being fixed.…”