Adaptive Control with Pseudo-Control Hedging for a Hypersonic Air Vehicle

For an air-breathing Hypersonic Air Vehicle (HAV) in cruise flight, the model is strongly nonlinear and coupled and includes uncertain parameters. A pseudo-linear model is obtained by input/output feedback linearization, and a model reference adaptive controller is applied to it. Single Hidden Layer (SHL) Perceptron Neural Network (NN) is used to cancel model error online because of its good nonlinear approximation. Considering that large-span maneuver or strong disturbance can cause control surface saturation, Pseudo-Control Hedging (PCH) is introduced. Simulation results show that the proposed adaptive control system has well tracking performance, strong robustness and anti-input-saturation characteristics.