A forward-backward PHD smoother for tracking multiple AUVs
Shangzhuo Zhang, Xiaomin Zhang, Yang Fan Yu · OCEANS 2016 - Shanghai · 2016
In this paper we address the problem of tracking multiple AUVs using a single underwater sensor. Using for this challenging scenario, in addition to target detections, in every scan the sensor returns clutter measurements. Standard target tracking in clutter most often uses the present position measurements only. we propose to use a forward-backward Probability Hypothesis Density (PHD) smoother and present a suitable implementation of the Gaussian Mixture PHD filter. The forward filtering is performed by Mahler's PHD recursion. The PHD backward smoothing recursion is derived using finite set statistics (FISST) and standard point process theory. The simulation results show that this algorithm has better tracking performance in high-density clutter underwater environment and better than Gaussian Mixture PHD (GM-PHD) filter.