Fault-Tolerance of a Transport Aircraft with Adaptive Control and Optimal Command Allocation

Abstract

How to achieve high performance and reliability against various unforeseen events, uncertainties and other changes in plant dynamics has been a very challenging issue for control system design in recent years. Reconfigurable flight controls aim to guarantee greater survivability in all the cases in which the systems to be controlled may be poorly modelled or the parameters of the systems may be subjected to large variations with respect to the operating environment. A suitable approach to the problem of flight control reconfiguration consists in redesigning its own structure and/or re-computing control gains in the case of unexpected events or large model and environmental uncertainties. A number of different approaches have been proposed and developed in the past years (Patton, 1997). In this chapter a Direct Adaptive Model Following (DAMF) algorithm has been used for reconfiguration purposes. It is possible to find in literature a great amount of proposed techniques to implement (Bodson & Groszkiewicz, 1997; Calise et al., 2001; Boskovic & Mehra, 2002; Kim et al., 2003; Tandale & Valasek, 2003). The Lyapunov theory described in (Kim et al., 2003; Tandale & Valasek, 2003) has very attractive features both in terms of effectiveness and implementation and it has been used to develop the fault-tolerant scheme described in this chapter. Another important matter in flight control reconfiguration is the Control Allocation (CA) problem. It concerns the possibility to exploit actuators redundancy with respect to the variables to be controlled in order to redistribute the control effort among the available control effectors. In this way the control commands needed to attain the desired moments can be computed even in presence of actuator failures, while also dealing with position and rate limits of the control effectors. A great amount of techniques for control allocation are available in literature (Virnig & Bodden, 2000; Enns, 1998; Buffington & Chandler, 1998; Durham & Bordignon, 1995; Burken et al., 2001). The technique used in this chapter is the one introduced by Harkegard (Harkegard, 2002) based on active set methods, which is very effective for real-time applications and converge in a finite number of steps. Therefore in this chapter a scheme of a fault-tolerant flight control system is proposed. It is composed by the core control laws, based on the DAMF technique, to achieve both robustness and reconfiguration capabilities, and the CA system, based on the active set method, to properly allocate the control effort on the healthy actuators. Numerical results of