Collaborative Estimation of State and Guidance Parameter for Interceptor Based on Variational Bayesian Technique

Abstract

In this paper, we study the state estimation problem with an unknown guidance parameter of the interceptor, where the guidance parameter is modeled by normal-gamma distribution. To solve the problem, we propose a variational Bayesian (VB) based collaborative estimation algorithm for state and parameters, where the joint posterior distribution (JPD) of state and guidance parameter is approximated by a free-form distribution. The proposed algorithm can be divided into two-stage iterative steps: in variational Bayesian expectation (VB-E) step, with guidance parameter fixed, the cubature Kalman filter (CKF) is employed to realize state estimation. In variational Bayesian maximum (VB-M) step, the statistical characteristics of the guidance parameter are then deduced with state fixed. The state and the guidance parameter can be effectively estimated by performing VB-E and VB-M steps recursively. Finally, we illustrate the effectiveness of the proposed algorithm by a collaborative estimation problem in the two-dimensional aerial engagement scenario.

Keywords

• Aerial recognition
• guidance parameter estimation
• collaborative estimation
• variational Bayesian
• cubature Kalman filter