An Augmentation Scheme for Fault Tolerant Control Using Integral Sliding Modes

Abstract

Abstract-In this paper a novel fault tolerant control allocation scheme is proposed, which has the capability to maintain closed-loop nominal performance in the case of faults/failures by effectively managing the actuator redundancy, and without reconfiguring the underlying control law. The proposed scheme relies on an a posteri approach, building on an existing state feedback controller designed using only the primary actuators. An integral sliding mode scheme is integrated with the existing controller to introduce fault tolerance. The proposed scheme uses the measured or estimated actuator effectiveness levels in order to redistribute the control signals to the healthy ones which allows a certain class of total actuator failures to be mitigated. The effectiveness of the proposed scheme is tested in simulation using a high fidelity nonlinear model of a transport aircraft model.

Output‐feedback sliding‐mode control for systems subjected to actuator and internal dynamics failures

“…This contrasts with other approaches which employ some effectiveness levels estimates of the actuators that could be provided by an FDI scheme [19]. The idea is similar to the dynamic control allocation introduced in [19], where the redundancy that is created by the use of primary and secondary actuators is essential to allow the distribution of the required control effort among the healthy actuators in the case of faults. As a result, the control distribution matrix, say K p (t), of the proposed scheme is time variant and uncertain because to actuator faults and other uncertainties.…”

“…In active FTC schemes, the mixer W is reconfigured to maintain the desired resultant action, for example, force in mechanical systems [41]. For instance, some estimate of the actuator gain matrix F(t) provided by an FDI subsystem can be used to compute the mixer matrix as in [17,19]. Here, we propose a passive approach based on unit vector modelreference SMC and an appropriate constant mixer matrix such that the closed-loop system stability and tracking performance are immune to some actuator faults and internal dynamics failures.…”

“…Then, subtracting (19) and (20) from the plant state equations (3) and (4) with control signals (11) and (21), the state error (x 1 := x 1 − x 1M ) and the output error (9) satisfẏ…”

“…Different faults can be considered in the design of a FTC scheme, such as sensor and actuator failures, for example. Most of the control algorithms are based on the assumption of full-state measurement [3,16,19]. As an alternative to this problem, some schemes have resort to the use of (high-order) sliding mode observers [2,4,6,8,20,25], which provide good performance for state and parameter estimation, even in the presence of measurement noise and unknown inputs.…”

“…On the other hand, passive FTC uses robust or adaptive control techniques to maintain the performance of the closed-loop system insensitive to certain faults [10]. Robust control [11][12][13], quantitative feedback theory [14], sliding-mode control (SMC) [15][16][17][18][19][20] and adaptive control [21][22][23][24] are some of the commonly used methods for passive FTC design.…”