A nonconvex cost optimization approach to tracking multiple targets

Abstract

The problem of tracking multiple targets in the presence of displacement noise and clutter is formulated as a nonconvex optimisation problem. The form of the suggested cost function is shown to be suitable for the graduated nonconvexity algorithm, which can be viewed as deterministic annealing. The method is first derived for the two-dimensional (spatial/temporal) case, and then generalized to the multidimensional case. The complexity grows linearly with the number of targets. Computer simulations show the performance with crossing trajectories.<>