Research On Maneuver Mode Analysis,Trajectory Tracking And Prediction Algorithm For Hypersonic Glide Vehicle

With excellent aerodynamic capability,hypersonic glide vehicles maneuver in non-ballistic mode and have the characteristics of enormous overload,high speed,and vast coverage area,which brings a significant challenge to the defense system.In defense,the guidance law design of the interceptor is based on the predicted set-forward position.Therefore,accurate trajectory prediction is the key to effectively intercepting hypersonic targets.At present,the research on trajectory prediction of hypersonic glide targets is still in the exploratory stage,and the understanding of its maneuver mode is not deep enough.Traditional trajectory tracking and prediction method cannot meet the requirements of hypersonic vehicle defense.Therefore,this dissertation focuses on maneuver characteristics and maneuver mode on multiple constraints in the gliding phase,joint estimation of state and parameter,as well as trajectory prediction methods.The main contents are as follows:(1)To explore the target maneuver mode,some classic maneuver modes and those under multiple constraints are studied,and the target’s state and characteristic parameters evolution patterns in each maneuver mode are analyzed.Firstly,maneuver characteristics,flight process constraints,and maneuverability are studied.Then,the principle of classic maneuver modes is studied and simulations on the evolution of state and characteristic parameters are performed.The maneuver modes,state,and characteristic parameter evolution patterns under multiple constraints are further studied based on various trajectory planning algorithms.In this way,the understanding of the maneuver mode in the actual flight scene is deepened and the characteristic parameters which are beneficial for constructing the maneuver modes are determined.Finally,the difficulties and feasibility of trajectory prediction in the gliding phase are analyzed,which is enlightening for maneuvering mode modeling,the joint state and parameter estimation algorithm design,and trajectory prediction method design.(2)Aiming at the maneuver mode modeling problem,a maneuver mode modeling method based on decomposition and superposition technique,as well as a parametric model is proposed by extracting and depicting the evolution rules of the time sequence of the target’s state and characteristic parameters.Firstly,the ensemble empirical mode decomposition method is used to extract the trend term and period term in the time series to avoid mutual interference between different mode data.Then,for the stationary trend term,the autoregressive model is used to accurately describe the evolution law of the trend term.For the non-stationary periodic term,the time-varying autoregressive model method is used to solve the periodic drift problem of the model by approximating the model’s time-varying parameters,and the modeling accuracy is significantly improved.The consistency between the model and the original periodic term sequence is verified from three aspects: time domain,frequency domain,and statistical characteristics.Finally,the parametric description of maneuver mode is realized by model superposition,which lays a theoretical foundation for maneuver mode identification and trajectory prediction.(3)A moving horizon estimation algorithm based on Carleman linearization is proposed for the joint estimation of state and maneuver characteristic parameters.Firstly,the maneuvering characteristic parameters are augmented into the state vector to construct the state/parameter joint estimation system.Then,to improve the estimation accuracy,the joint estimation problem of state/parameter is transformed into a constrained optimization problem in a sliding window based on the moving horizon estimation principle and the prior constraints of state and characteristic parameters.The time cost of continuously solving optimization problems is very high,especially when the inequality constraint is introduced,the time cost will increase greatly.Based on the Carleman linearization method and combining the characteristics of the moving horizon estimation,the original estimation problem is converted into the constrained optimization problem of a linear system,while guaranteeing the estimation precision and reducing the time cost.Aiming at the arrival cost updating problem,the approximate form of arrival cost is derived and the CKF arrival cost updating algorithm is designed,which effectively improves the arrival cost updating accuracy.Finally,the stability of the algorithm is analyzed and the stability conditions are derived.(4)Aiming at the trajectory prediction problem,a short-term trajectory prediction method based on maneuver mode online identification and a longitudinal long-term motion trend prediction method are proposed,as well as a maneuver detection decision mechanism is developed to reduce the impact of the maneuver mode mutation uncertainty and effectively improve the trajectory prediction accuracy.As for short-term trajectory prediction,the trajectory prediction model is constructed with a combination of the dynamic model and the parametric model of identified maneuver mode to realize trajectory extrapolation prediction.In the case of no maneuver mode mutation,the method has high prediction accuracy.To deal with the lateral maneuver mode mutation,a maneuver detection decision mechanism is developed using the maneuver detection method based on higher-order cumulants,which improves trajectory prediction accuracy efficiently.Aiming at the problem of longitudinal long-term motion trend prediction,according to the evolution law of longitudinal trajectory,the proposed maneuver mode parametric modeling method is used to construct the longitudinal trajectory prediction model,to realize the longitudinal longterm motion trend prediction.The algorithms above realize the short-term accurate prediction and the longitudinal long-term trend prediction of gliding trajectory,as well as provide a theoretical basis for the guidance law design of interceptor,interception scheme design,and defense resource allocation.