Adaptive Flight Control Design for Fixed-wing UAV based on Neural Network

This paper presents an adaptive flight control approach based on dynamic inversion and neural network with sliding mode technique for a fixed-wing unmanned aerial vehicle (UAV). A nonlinear dynamic model which has been decoupled into three independent single-input single-output channels is investigated. Unknown certainties of the aircraft dynamic are estimated by the neural network, which adaptively provides online parameter regulation, and the close stability analysis is also derived by using Lyapunov theory. Simulation and experiment results indicate that the proposed control scheme can perform well in tracking problem.