An underwater bearing-only multi-target tracking approach based on enhanced Kalman filter

Abstract

This paper presents an enhanced Kalman filter, including the multi-track gate method and the autoregression (AR) model, for underwater bearing-only multi-target tracking. The single-double side constant false alarm rate (SD-CFAR) method is firstly proposed for crossing target detection, and multi-track gate method and autoregression (AR) model is then used to enhance the traditional Kalman filter to complete automatic track initialization, crossing trace tracking and track interruption prediction. The results of simulation study verify the effectiveness of the presented approach for bearing-only multi-target tracking and indicate that this approach is more beneficial than traditional CFAR and Kalman filter.