Dynamic surface control based on neural network for an air-breathing hypersonic vehicle

In this paper, an adaptive dynamic surface control approach is presented for the longitudinal motion of an air‐breathing hypersonic vehicle. Fully tuned radial basis function neural network that regulates weights, width, and center of Gaussian function simultaneously is developed to estimate aerodynamic uncertainties and atmospheric disturbances. The nonlinear control law is subsequently designed by dynamic surface control approach for the vehicle model converted into strict block feedback form by input–output linearization method. Simulation results show that the velocity can be successfully tracked over a large range from Mach 11 to Mach 12 and an altitude range from 26 to 30km. The presented approach has perfect ability of restraining unknown and time‐varying nonlinear dynamics during flight.