Backstepping Controller Design Using Neural Networks for a Fighter Aircraft

This paper presents an on-line neural network based adaptive controller by using the backstepping technique for flight control systems in which command following of the angle of attack, sideslip angle, and roll angle about the velocity vector are property tracked. The network is trained in such a way that the overall system becomes robust against the uncertainties due to aerodynamic coefficients variations or sudden change in aircraft dynamics while the closed-loop stability is guaranteed. Finally, nonlinear six-degree-of-freedom simulation results for an F-18 aircraft model are presented to demonstrate the effectiveness of the proposed control law.