An Adaptive Fault Management (AFM) System for Resilient Flight Control

Jovan Boskovic, Joshua Redding, Nathan Knoebel
2009 AIAA Guidance, Navigation, and Control Conference   unpublished
One of the major problems that arises in the context of upset prevention and recovery in flight control systems is the availability of the information about the new aircraft dynamics post-failure or post-damage. For that reason it is important to have an on-board system that will rapidly and accurately detect and identify a failure (or failures), or the location and extent of structural damage. However, such systems are commonly prone to false alarms. If a false failure information is used in
more » ... e feedback control law, this can lead to substantial performance deterioration and even instability of the closed-loop system. One possible solution to this problem is Self Diagnostics (SD) -a procedure where probing signals are injected into the system to try to identify failure or damage-related parameters. Such a procedure needs to be implemented with care since it can potentially excite aircraft structural modes. In addition, self diagnostics can induce an oscillatory response of the aircraft states while not still being insufficient to accurately identify the parameters, particularly if their number is large. Hence such an approach is effective only if the number of failure-related parameters is small. Knowledge of the post-fault parameter values is important for: (i) calculation of available control authority immediately following an upset and control reconfiguration, (ii) computation of achievable trim points; and (iii) calculation of new control constraints. In this paper we present and discuss our Fast on-Line Actuator Reconfiguration Enhancement (FLARE) Systemthat is based on decentralized detection and identification of flight control actuator failures, and results in a relatively small number of parameters that need to be identified on-line. FLARE was recently augmented by the self-diagnostics module, resulting in an effective system that accurately estimates the parameters such that the effect of the self diagnostic signals is not felt by the aircraft states. Hence the possibility of exciting the structural modes is minimized. The FLARE system with self diagnostics (FLARE-SD) is described in detail, as well as the areas where such knowledge can be effectively used to increase the pilot's awareness and improve flight safety. The features of the FLARE-SD system are illustrated through simulations of F/A-18 aircraft dynamics under actuator failures. *
doi:10.2514/6.2009-6263 fatcat:giuogsmlpva3hcwsdy3xx2v3iy