Reconfigurable flight control system design using adaptive neural networks

An adaptive controller design method based on neural network is proposed for reconfigurable flight control systems in the presence of variations in aerodynamic coefficients or control effectiveness deficiencies caused by control surface damage. The neural network based adaptive nonlinear controller is developed by using the backstepping technique for command following of the angle of attack, sideslip angle, and bank angle. On-line learning neural networks are implemented to compensate the control effectiveness decrease and guarantee the robustness to the uncertainties due to aerodynamic coefficients variations. The main feature of the proposed controller is that the adaptive controller is designed by assuming that all of the nonlinear functions of the system have uncertainties, whereas most of the previous works assume that only some of the nonlinear functions are unknown. Neural networks learn through the weight update rules that are derived from the Lyapunov control theory. The closed-loop stability of the error states is also investigated. A nonlinear dynamic model of a high performance aircraft is used to demonstrate the effectiveness of the proposed control law.