Adaptive Neural Network Tracking Control Algorithm For Air-breathing Hypersonic Vehicle

This paper mainly studies the tracking control of a kind of air-breathing hypersonic vehicle.Because the dynamic model of this kind of vehicle has the characteristics of high nonlinearity,complex coupling,and parameter uncertainty.However,traditional control methods cannot solve the real-time and accuracy problems in tracking control research.Therefore,the flight control design of the existing air-breathing hypersonic vehicle is still an open challenge.For this kind of tracking control problem,this paper proposes two new controller design methods,adaptive neural network control algorithm,and adaptive neural network control method with dead zone error.Simulation experiments verify that the proposed methods can speed up the tracking control of the vehicle and improve the accuracy of the tracking control.The specific work content is as follows:1.A neural network control algorithm is designed for a simplified control model of hypersonic vehicles that includes flight path,flight altitude,and angle of attack.The algorithm uses the SDU matrix to decompose the gain matrix in the simplified model and eliminate the strong coupling between the control input and the control output,and reduce the complex coupling of the hypersonic vehicle model itself.At the same time,the set of unknown functions in the model after approximate decomposition is applied to the neural network to solve the parameter uncertainty.Through the Lyapunov stability analysis,it is verified that the control system can achieve fasttracking of the target.2.To improve the tracking performance of the multiple-input multiple-output(MIMO)rectangular nonlinear system,an adaptive neural network control method with dead zone error is proposed for the tracking control of air-breathing hypersonic vehicle.And to improve the adaptive neural network control algorithm,an adaptive neural network with a dead zone error update rate is introduced to generate control actions.The control method can be better applied to the air-breathing hypersonic vehicle system with parameter uncertainty and interference and can track the desired flight trajectory,flight altitude,and angle of attack well.Through Lyapunov stability analysis and PID simulation experiments,the stability and tracking performance of the closed-loop system is verified.