Skip to Main Content
This study is devoted to the problem of state estimation of discrete-time stochastic systems with Markov switching parameters. Three improved interacting multiple model (IMM) algorithms for manoeuvring target tracking are presented, in which the filter outputs are combined based on three optimal multi-model fusion criterions weighted by scalars, diagonal matrices and general matrices, respectively. The proposed algorithms can receive the optimal state estimations of target in the linear minimum variance sense. It is proved that the traces of variance matrices of tracking errors in three proposed algorithms are less than the trace in the classical IMM algorithm. Extensive Monte Carlo simulations verify that the proposed algorithms are effective and have an absolute advantage in the velocity estimation. In particular, one of the proposed algorithms is obviously better than the IMM algorithm in accuracy and elapsed time and, therefore, can be a competitive alternative to the classical IMM algorithm for the tracking of manoeuvring target in real time.