Positioning And Trajectory Prediction Of Hypersonic Vehicle

In recent years,with the rapid development of Aerospace technology in major powers,near-space hypersonic vehicles have been increasingly prepared for their unique advantages such as fast speed,good maneuverability,strong penetration capabilities,fast precision attack and long-distance rapid delivery.These characteristics make the detection and interception of this type of target face great challenges,and pose a major threat to the national defense security.Therefore,it is of great significance to study the positioning and trajectory prediction of hypersonic vehicles.However,due to the introduction of external forces in the flight process,the aerodynamic analysis is complex,and the flight trajectory is separated from the inertial trajectory,resulting in its state is difficult to predict.In addition,the flight speed is fast,the flight trajectory is complex and changeable,and the uncertainty of the prediction model is caused by the difficulty in determining the motion model.In this paper,the theoretical analysis and simulation experiments are combined to study the positioning and trajectory prediction of the hypersonic vehicle based on the research framework of positioning before prediction,combined with the simulation data of the hypersonic vehicle,as follows:Firstly,the ballistic characteristics of the target vehicle are understood and analyzed,and the longitudinal motion equations of the vehicle are obtained.The motion equations are solved by using the fourth-order Runge Kutta method.The longitude,latitude and height obtained by solving the equation are transformed into the space rectangular coordinates under the inertial coordinate system,which provides the data basis for the positioning and trajectory prediction of the vehicle.Secondly,aiming at the problems that ground radars are difficult to detect and warn vehicle with maneuvering,which is easily affected by the curvature of the earth,the target location method based on the double star observation system is studied.Create the ephemeris file of the 3D coordinate data of the vehicle and import it into STK.The trajectory model of the "jump glide" hypersonic vehicle is established.The target observation scene based on the binary star system is built.The observation vector model between the platform and the target under the earth inertial system is constructed,using the new geometric positioning algorithm of the target by the binary star observation system,the target positioning modeland positioning error model in the inertial system are deduced,and the positioning error distribution is obtained by Monte Carlo method,which realizes the supersonic vehicle Positioning.The simulation results show that the algorithm can control the target positioning error within 10%,and the average positioning accuracy of the target position is below the kilometer level,which proves the effectiveness of the algorithm.Finally,in this paper,two trajectory prediction methods,dynamic model and kinematic model,are used to solve the problems of strong maneuverability,complex motion characteristics and different mastery of prior knowledge of the target.The former obtains a simplified dynamics model by analyzing the force of the hypersonic vehicle during the glide phase and reasonably simplifying the conditions.The analytical form of the trajectory is obtained by solving the differential equation,and the long-time trajectory prediction is realized.According to the characteristics of periodic jump motion of non-ballistic hypersonic vehicle,combined with singer motion model,the latter use the improved UKF algorithm track and estimate the target motion state.The algorithm enhances the tracking performance of the filter and reduces the initial value error of the state filter and the influence of the disturbance error of the state equation on the filtering result by introducing a regulating factor.Simulation results show that the improved adaptive UKF algorithm has better tracking accuracy and stability than the traditional algorithm.By using the state estimation of the filtered target and the acceleration fitting curve of DPA + LA,the trajectory of the target vehicle is further recursive.The simulation results show that the prediction error of this method is within 1km.