Guidance And Trajectory Prediction Of Near Space Interceptor Missiles

In recent years,near space hypersonic vehicle technology has continuously developed,and the demand for near-space offensive and defensive countermeasures technology has become more and more urgent.From the perspective of the defender,this thesis studies the design of the guidance law and guidance filter of the anti-near-spacehypersonic-target interceptor missiles;from the perspective of the offensive side,this thesis also studies the guidance law identification and trajectory prediction of the near space interceptor missiles.At low altitude,the dynamics of an actual interceptor missile's autopilot are usually more complex,which are a second-order system,and have two zeros.For tail-controlled missiles,there will also be a zero in the right half plane.This will decrease the performance of guidance law.In order to solve this problem,this thesis proposes a differential game guidance law that considers the autopilot dynamics of two poles and two zeros.In this case,the system model can become very complicated.In order to deal with such a situation,it is assumed that the missile and the target make decisions separately,which can simplify the model and obtain an analytical solution by the optimal control method.We know that the optimal guidance law is equivalent to the augmented proportional guidance law with a guidance coefficient of 3.If the dynamics of the autopilot are not taken into account,the guidance law proposed here can also be converted into a similar form.Finally,the effect of the guidance law is verified by numerical simulations.In the simulations,the augmented proportional guidance law is compared with guidance laws considering different dynamics model of the autopilot.It is found that the guidance laws considering the dynamics of the autopilot can obtain better interception performance,and that the guidance law considering the dynamics of the autopilot as a two-pole and two-zero system works best.In many near space combat conditions,interceptor missiles need to release a Kinetic Kill Vehicle(KKV),which can kill the target through the terminal guidance phase.How to obtain an accurate estimation of the line-of-sight(LOS)angular rate through the measurement of the seeker in the terminal guidance phase is an important research topic.In KKV's guidance system,the seeker usually adopts strapdown or semi-strapdown structure.In order to intercept near space targets,KKV generates the required lateral thrust by switching on and off the pulse thrusters.In fact,this process can cause a large impact vibration on the missile body,which deteriorates the measurement environment of the LOS angle.There is not much research on this issue by now.One way to deal with this problem is to use the impact vibration and the LOS angular rate altogether as the estimated vector of the Kalman filter.First,the observability of the target-missile relative motion system considering impact vibration is demonstrated,based on which,an augmented-state Kalman filter is designed.In order to improve the precision and speed of the filter,two specific filtering techniques,continuous-discrete and optimal two-stage,are introduced.The simulation results show that the estimation of the impact vibration can improve the accuracy of the LOS angular rate extraction and improve the interception effect.The introduction of continuous-discrete and optimal two-stage further improves the filter's performance.The continuous development of the near space interceptor missiles has also brought enormous challenges to the offensive side,making the near space offensive side vehicles take further active measurements.The offensive side hopes to predict the future position of the interceptor missile via the information of its guidance law,and then develop strategies to avoid the interception based on the prediction.In this context,this thesis studies the problem of guidance law identification and trajectory prediction of the near space interceptor missiles.There are many types of guidance laws developed now.However,they can be all approximated by the typical proportional guidance law.Firstly,the target-missile relative motion model under proportional guidance law is established,and the observability of the system is analyzed by nonlinear system theory.After that,by applying two nonlinear filtering algorithms,extended Kalman filtering and unscented Kalman filtering,nonlinear filters for the problem of guidance law identification are designed.The two nonlinear filtering methods are verified and compared by numerical simulations.Finally,the problem of trajectory prediction is studied.The calculation model for trajectory prediction of the interceptor missile is established.The prediction effect is verified by numerical simulations and the error analysis of the simulation results is made.The actual guidance situation of interceptor missiles can be more complicated.This thesis next studies one of them: the proportional guidance law appears to have acceleration command saturation.This situation usually occurs when the initial heading error is large.Under the proportional guidance law,the load command be proportional to the LOS angular rate,so a large acceleration command is generated at the initial stage and thus saturated.In this case,the relationship of the proportional guidance law is actually not satisfied,and the single model filtering method used in the previous is no longer available.Then we discuss the problem of using the multiple model filtering method to identify the proportional guidance law with saturation.Firstly,based on the original proportional guidance law model,the model under saturation is added and its observability is analyzed.After that,the interactive multiple model(IMM)algorithm is used to do the filtering task.The IMM algorithm's ability to switch models is very suitable for the actual situation of this problem,thus it is a good processing method.The specific inplementation of the algorithm is discussed in detail in the thesis.The effect is verified by numeric simulation,and the results are then analyzed.