Adaptively tracking hypersonic gliding vehicles

Abstract

The motion model plays a key role in tracking hypersonic gliding vehicles (HGVs). The process noise covariance matrix is used to scale the errors of the motion model, and its design has significant influence on the tracking accuracy. In this paper, a dynamics model with an adaptive process noise covariance matrix is proposed for tracking HGVs. The dynamics model is described by a nonlinear stochastic differential equation (SDE), and its corresponding process noise covariance matrix is calculated by adopting the modified Gauss-Legendre scheme. Furthermore, an adaptive adjustment method for process noise covariances of the augmented states is designed by analyzing the variation characteristics of the model errors. Specifically, the process noise covariances are adjusted by an adaptive factor: the square of the ratio of the estimated dynamic pressure to a prior dynamic pressure. Simulation results show that the proposed model can provide higher tracking accuracy than the dynamics model with fixed process noise covariances.