Research On Target Tracking Method Based On Interactive Multiple Model

With the development of modern information technology and control technology,anti-ship missile with its changeable mobile form and powerful penetration capability poses a great threat to the maritime military equipment of various countries,therefore,the study of high-precision filter tracking algorithm,real-time monitoring and tracking of anti-ship missiles,the improvement of anti-missile capability of maritime military equipment has an important significance.This paper is based on the interactive multi-model algorithm for research,the main contributions are divided into the following aspects:Firstly,several common target tracking models and nonlinear filtering algorithms are studied.In the tracking model,the focus is on the "snake-like" motion model of anti-ship missiles.In terms of filtering algorithm,the detailed theoretical derivation and applicability analysis of UKF algorithm,CKF algorithm and 5CKF algorithm are carried out.In view of the disadvantages of low precision of the filtering algorithm,the Sage-Husa filter theory is studied,and a five-degree cubature Kalman filter algorithm based on error covariance adaptive is proposed in combination with the 5CKF algorithm,it has better tracking accuracy and convergence speed.Secondly,aiming at the error problem caused by the fixed Markov probability transition matrix in the interactive multi model algorithm,this paper proposes an interactive multiple model adaptive five-degree cubature Kalman filter algorithm(AIMMA5CKF)based on Markov parameter adaptive,which is a method based on the posteriori information correction.The algorithm uses the defined error compression ratio to realize the adaptive adjustment of markov probability transfer matrix,which makes the matching model information increase and the unmatched model information decrease in the process of model switching,so as to reduce the tracking error.Finally,in view of the error problem caused by the model probability adjustment speed in the traditional interactive multi-model algorithm,this paper proposes an interactive multiple model adaptive five-degree cubature Kalman filter algorithm(FAIMMA5CKF)based on fuzzy adaptive.The algorithm first uses the high-precision A5 CKF algorithm to realize the adaptive adjustment of error covariance,and then uses the fuzzy control algorithm to realize the adaptive update of the model probability,so as to speed up the update speed of the model probability and further improve the tracking performance of the system.