A Multi-point Converted Measurement Algorithm for Underwater Maneuvering Target Tracking with Gaussian Mixture Noises

In this paper, we propose a multi-point simultaneous converted measurement filter to solve the issue of outlier-corrupted noise in maneuvering target tracking. Optimal parameters for the student’s t distribution are selected by KLD criterion to approximate the statistical characteristics of Gaussian mixture measurement noise. The performance degradation of measurement outliers is alleviated by multi-point filtering, leading to higher model judgment in the IMM algorithm. Simulation results show that this method is better than the existing methods in terms of tracking accuracy and consistency.