Intelligent Morphing Decision of Hypersonic Aircraft Based on Neural Network

Aiming at the autonomous morphing decision problem of hypersonic morphing glide aircraft, an intelligent morphing decision method based on neural network is proposed in this paper. Firstly, establishing a dynamic model of hypersonic glide aircraft with continuously variable sweep angle. Then, considering the heat flux constraint during gliding, generating the trajectory sample library by Gaussian pseudo-spectrum method, and the flight state is taken as the network input and the sweep angle command as the network output for neural network training, which realizes the intelligent decision of sweep angle based on flight state during gliding. Finally, applying designed intelligent decision maker to trajectory optimization. The simulation results show that the sweep angle can still be intelligently decided according to the current flight state to improve the range after the glide trajectory is disturbed, which proves the robustness of the intelligent decision-making instrument.