Performance of PMHT based algorithms for underwater target tracking

Target tracking in underwater environment is a challenging problem because of intense reverberation and clutter. At the center of this problem lies the measurement-to-target association problem. Conventional tracking filters assumes that only one measurement can be observed for a target, while PMHT assumes that more than one measurement can be observed for a target at each sampling time. Conventional approaches produce sub-optimal while PMHT produces optimal solutions for underwater target tracking with these assumptions. Another point that makes PMHT important is its flexible and easily modifiable model. The PMHT derivatives are introduced with the methods produced to solve the known problems of PMHT such as nonadaptivity, narcissism, over-hospitality to clutter. In this paper comparative performance results for PMHT and some of its derivatives will be studied for the underwater tracking application.