An Interacting Multiple Model Smoothing Algorithm for Nonlinear Maneuvering Target Tracking

In order to improve the accuracy of nonlinear maneuvering target tracking system, this paper proposes a novel IMM smoothing algorithm, which can be used in two types of IMM smoothing by changing the delay time of IMM smoother, for nonlinear maneuvering target tracking. The IMM method combined with smoothing algorithm provides more accurate maneuvering target state estimation. The novel IMM smoothing is consisted of forward IMM filtering and backward IMM smoothing. The forward IMM filtering is a standard IMM filter, while the backward IMM smoothing utilizes the updated value and predicted value of the forward IMM filtering to get the smoothing state of the maneuvering target. Simulation results show that the novel IMM smoothing algorithm has better performance compared with the other algorithms.