Aircraft reconfiguration using neural generalized predictive control

Abstract

The objective of this paper is to report the preliminary results from the research being conducted in reconfigurable flight control. This paper highlights the Neural Generalized Predictive Control algorithm, which is capable of real-time control law reconfiguration, model adaptation, and the ability to identify failures in control effectiveness. This paper presents results for a full mission six degree-of-freedom conceptual commercial transport aircraft simulation where the elevator is frozen during the flight and the algorithm reconfigures to use symmetric aileron deflections to control pitch rate, thereby stabilizing the aircraft. The neural network is activated to learn the changed dynamics of having frozen elevators and performance is improved.