Research On Trajectory Prediction And Advanced Interception Methods Of The Near-space Hypersonic Vehicile

The rapid development of near-space hypersonic vehicles has caused a huge impact on the existing missile defense system.To effectively intercept the near-space targets with fast speed,high and continuous maneuvering ability,this paper has studied the state estimation and prediction of the target,the design of the optimal interception guidance law and the design of the multi-missile cooperative interception guidance law.Firstly,this paper has established a near-space vehicle maneuvering model for state estimation.The maneuvering model fully considers the characteristic that the near-space vehicle is maneuvering by aerodynamic.By introducing three state variables related to aerodynamic coefficients,the established nonlinear maneuvering model can more accurately and quickly respond to changes in target acceleration.The local observability theory of nonlinear systems is applied to prove the observability of the model,and then the target states are estimated using the extended Kalman filter.Considering the blind spot problem of radar detection devices in the detection of near-space targets,to increase the detection time and provide more sufficient time for subsequent interception,a spacebased infrared and ground-based radar cooperative detection scheme has been proposed.The cooperative detection scheme ensures that the state estimation filter can have a good initial value after the target enters the radar field of view,thereby realizing the rapid convergence of the filter;the cooperative detection scheme can also independently determine the infrared detection device with smaller measurement error,and obtain a measurement of the target position accordingly.According to the typical maneuvering mode of the near-space vehicle,an iterative prediction method of the target future states has been realized by polynomial fitting or treating the introduced aerodynamic related state as a constant value.The prediction method can be realized without prior information such as target mass,aerodynamic coefficient,and reference area.In order to make full use of the obtained predictions of the target states,an energyoptimized guidance law considering the predicted information has been designed in this paper.This guidance law can effectively reduce the energy consumption of missiles that use direct force for lateral maneuvering when intercepting maneuvering targets.To improve the robustness of the optimal guidance law and deal with the prediction errors,an adaptive sliding mode control part has been designed,and the optimal sliding mode guidance law has been proposed.The designed adaptive sliding mode control part can also solve the problem of input saturation constraint of the missile.Since the predicted information of the target in the future is used in the design of the guidance law,the overload command value of the missile at the end of the interception can be effectively reduced when the prediction result is ideal,so that the missile has more sufficient available overload to cope with the target evasion maneuver.Furthermore,a model prediction guidance law based on the linear Gaussian pseudospectral method has been proposed.This guidance law takes the state and control input of the optimal sliding mode guidance law as the nominal state and nominal input.Since there is an analytical solution to the linear quadratic optimal control problem,by performing local linearization at the nominal state,the guidance command can be quickly obtained only by solving the linear matrix equation.It solves the shortcomings of high computational complexity and poor real-time performance of traditional model predictive control.Since model predictive guidance law does not require decoupling the three-dimensional guidance problem into a two-dimensional plane problem,the decoupling error is avoided,and the requirement for the selection of the missile interception coordinate system is weakened.To successfully achieve the simultaneous interception of a near-space target by multiple intercepting missiles,a closed-loop analytical method for estimating the timeto-go of missiles using proportional navigation law,which has high accuracy for stationary and maneuvering targets,has been proposed in this paper.By approximating the missile flight process as a constant lift-drag ratio stage and a constant drag coefficient stage,the influence of drag can be considered when calculating the remaining flight time of the missile.In this paper,the time-to-go estimation method for intercepting a stationary target is given first.Then,a maneuvering target is transformed into a stationary target by reasonably setting a virtual target related to the motion state of the target,and a preliminary estimation of remaining flight time for maneuvering targets is obtained.To improve the estimation accuracy,an estimation compensation term is designed on the basis of the preliminary estimation.A time-coordinated cooperative interception for a near-space maneuvering target has been realized by using the time-to-go estimation.On this basis,a time-angle constrained cooperative guidance law with control mode switching capability is designed,which can realize the joint control of the missile attack time and attack angle.Considering that the maneuverability of a near-space hypersonic target may be higher than that of an intercepting missile,this paper has designed a cooperative interception guidance law based on coverage theory,which used the missile’s quantitative advantage to improve the interception success rate.The guidance law introduces the concept of coverage probability.For the whole interceptor missiles,the guidance law aims to maximize the coverage probability.Then,the gradient descent method is used to solve the solution online to obtain the expected flight-path angle command of each missile that maximizes the coverage probability under the current relative states of the intercepting missiles and the target.For a specific interceptor missile,the cooperative guidance law transforms the guidance problem into a flight path angle command tracking problem,which can be realized by finite-time state-dependent Riccati equation(FT-SDRE)technique.