Nonlinear discrete-time reconfigurable flight control law using neural networks

Abstract

A neural-network-based adaptive reconfigurable flight controller is presented for a class of discrete-time nonlinear systems. The objective of the controller is to make the angle of attack, sideslip angle, and bank angle follow a given desired trajectory in the presence of control surface damage and aerodynamic uncertainties. The adaptive discrete-time nonlinear controller is developed using the backstepping technique and feedback linearization. Feedforward multilayer neural networks (NNs) are augmented to guarantee consistent performance when the effectiveness of the control decreases due to control surface damage. NNs learn through the recursive weight update rules that are derived from the discrete-time version of Lyapunov control theory. The boundness property of the error states and NN weight estimation errors is also investigated by the discrete-time Lyapunov analysis. The effectiveness of the proposed control law is demonstrated by applying it to a nonlinear dynamic model of the high-performance aircraft.