Stabilized MPC formulations for robust reconfigurable flight control

Model predictive control (MPC) solves an optimization at every sampling instance to achieve commanded set points and control objectives subject to constraints on control inputs and system states. Such online optimization can incorporate many important factors to enable the control of systems subject to faults, changing dynamics, changing control objectives, failed control inputs and large disturbances. Reconfigurable flight control is one such application where the capabilities and flexibility of optimization-based control methods can be fully utilized and exploited. This paper presents formulations and experimental evaluations of various MPC schemes applied to a realistic full envelope non-linear model of a fighter aircraft. Investigations are carried out by exploring a variety of scenarios of fault and disturbance combinations along with modified and robust formulations of online constrained optimization.