Research On Trajectory Optimization And Control Approach For Hypersonic Vehicle Under Uncertainty

Hypersonic aircrafts attract worldwide attention due to advantages such as the long flight distance and strong maneuvability.However,wide flight envelope,complicated aerodynamic coupling,severe elastic vibration and unavoidable aerodynamic parameter perturbation make the dynamics model of hypersonic vehicles difficult to describe accurately.This paper takes hypersonic vehicles as the object of study and focuses on improving the robustness of the flight trajectory and attitude control systems.The dynamics model,flight trajectory characteristics and complex coupling systems are studied as well as the influence of uncertainty and external disturbance.The main research results are as follows:Considering the rotation and flattening of the earth,the reentry process of the hypersonic vehicle is analyzed,and the hypersonic reentry dynamics model in the half-speed coordinate system is given.At the same time,process constraints and geometrical constraints such as no-fly zones are studied.Then a multi-constrained reentry flight trajectory optimization model is established to meet the task demand.The hypersonic trajectory optimization model is solved by Gauss pseudospectral method,and the optimal reentry trajectory is obtained under multiple constraints.Based on the integral results of the discrete control varibles,Gauss pseudospectral method is proved feasible to solve the trajectory optimization problem,which lays the foundation for following trajectory optimization researches.Aiming at the perturbation of aerodynamic parameters in hypersonic reentry,a new method of uncertainty expansion based on generalized polynomial chaos theory is proposed.Compared with the traditional Monte Carlo simulation method,the generalized polynomial chaos method has high coincidence degree,but the number of sampling points can be greatly reduced.Thus the calculation efficiency is improved remarkably.By analyzing the statistical properties of each state variable and non-linear constraint function in trajectory optimization process,a reentry trajectory robust optimization strategy is established based on generalized polynomial chaos method.The strategy can effectively reduce the sensitivity of the optimized trajectory to the aerodynamic parameter peturbation,and shows strong robust stability.Combining the polynomial chaos method with Kriging method,a hybrid surrogate model optimization method is proposed.Global characteristics of the model are described by the polynomial chaos method,and local characteristics are approximated by the Kriging method.The cross-validation method is used to compare the prediction error of the hybrid algorithm with each single method.The results of the optimization are also compared.The simulation results show that the hybrid trajectory optimization method based on polynomial chaos and Kriging method has satisfactory performance,and shows obvious superiority.The hybrid algorithm can solve the trajectory optimization problems with multiple complex constrains and generate an optimal trajectory in a short time based on the initial reentry states obtained online,which avoids the recalculation of the original optimization model and saves a lot of calculating time online.The elastic mode vibration is regarded as the external disturbance of the hypersonic vehicle,and the dynamic characteristics of the vehicle are approximately described.Then,the control design model of the hypersonic vehicle can be established.The backstepping method is used to overcome the shortcomings of the existing methods which need to linearize the model of the aircraft.Thus,the design process of the control system is simplified.The lumped disturbance estimated by the nonlinear disturbance observer is introduced into the whole system to guarantee the robustness.The stability is demonstrated by Lyapunov method,which means the tracking errors of the speed and altitude converge in a finite time.Considering the uncertainties,perturbation of parameters and high-frequency noise,the control method proposed in this paper is compared with nonsingular terminal sliding mode control method and classical backstepping method,which demonstrates the effectiveness and advantages of the proposed method.In this paper,trajectory optimization method for hypersonic vehicles and design method of control systems under the uncertainty are preliminarily studied.The results of this dissertation can be used as reference for other novel spacecrafts.