Research on Extended Target-Tracking Algorithms of Sea Surface Navigation Radar

Abstract

To solve the problem of false tracks generated by breakdowns and clutter in point-target tracking in polar coordinates, a fusion tracking algorithm based on a converted measurement Kalman filter and random matrix expansion is proposed. The converted measurement Kalman filter (CMKF) transforms the polar coordinate data of the target at the current time into Cartesian coordinates without bias. Based on linear measurements and states, the position of the extended target and the group target was predicted and updated by using a random matrix, and its track was drawn by combining the nearest neighbors to realize the tracking of the size, shape and azimuth of the extended target. Compared with point-target tracking, the accuracy of extended multi-target tracking was increased by 45.8% based on data measured using NAVICO navigation radar aboard ships at sea. The experimental results showed that the improved method in this paper could effectively reduce the interference of clutter on target tracking and provide more information about the target motion features.