Design of Reconfigurable Flight Control System Using Adaptive Sliding Mode Control: Actuator Fault

Abstract

This article presents the reconfigurable flight controller using an adaptive sliding mode control scheme for actuator fault case. Sliding mode controller, which has good performance for the systems with various uncertainties, is used to deal with the actuator faults. Actuator fault can be considered as a disturbance or an unexpected parameter change, which degrades the system performance and may destabilize the system. In this study, the adaptive sliding mode control technique is adopted to compensate the effects of the disturbance generated by actuator faults. Lyapunov stability theory is used to derive the adaptive rule, and the closed-loop system stability analysis is performed. To demonstrate the effectiveness of the proposed controller, numerical simulation is performed for aircraft having redundant control surfaces.

Fault tolerant control using sliding modes with on-line control allocation

“…Here it is assumed that the matrix B 2 represents the dominant contribution of the control action on the system, while B 1 generally will have elements of small magnitude compared with B 2 . Compared to the work in [16] where it is assumed that B 1 = 0, here B 1 = 0 will be considered explicitly in the controller design and in the stability analysis. It will be assumed without loss of generality that the states of the system in (1) have been transformed so that B 2 B T 2 = I l and therefore B 2 = 1.…”

“…The work in [17,21] provides practical examples of the combination of sliding mode control (SMC) and CA for FTC. The work by Shin et al [16] uses control allocation ideas, but formulates the problem from an adaptive controller point of view. However neither of these papers provide a detailed stability analysis and discuss sliding mode controller design issues when using control allocation.…”

Fault Tolerant Control Using Sliding Modes with On-Line Control Allocation

“…In this paper, information about K(t) will be incorporated into the allocation algorithm . In most CA strategies, the control signal is distributed equally among all the actuators [16] or distributed based on the limits (position and rate) of the actuators [8].…”

“…where the positive scalar γ 0 is defined in Proposition 1, the positive scalar γ 1 is defined in (16).…”

“…where γ 0 , γ 1 and γ 2 are defined in (12), (16) and (19), and ∆ max is a fixed positive scalar. Then choosing…”