Bearing only tracking for maneuver target using nonlinear second-order Markov model

Abstract

In this paper, in addition to investigation and analyzing the dynamic model of a maneuver target, a new method based on the Interaction Multiple Model (IMM) method is presented to solve the tracking problem in presence of measurement noise. In this procedure, two models are used along with an extended Kalman filter for each model, for estimation of the states related to stochastic target model. To this end, a specific weight is calculated adaptively for each model and the final estimation of the target is obtained from the weighted sum of the modes related to each model. In this paper, second order Markov models are used to better describe the system behavior which leads to a decrease in the number of required motion models. This means that the previous two models are used to decide on the next model, and a much better algorithm is provided than the first-order IMM algorithm