A nonconvex cost optimization approach to tracking multiple targets

K. Rose, E. Gurewitz, G. Fox
IEEE International Workshop on Intelligent Robots and Systems, Towards a New Frontier of Applications  
A b s t r a c t The problem of tracking multiple targets in the prqence of displacement noise and clutter is formulated as a nonconvex optimization problem. The form of the suggested cost function is shown to be suitable for the Graduated Non-Convexity 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.
more » ... omputer simulations show the performance with crossing trajectories. U
doi:10.1109/iros.1990.262365 dblp:conf/iros/RoseGF90 fatcat:u2p4ftbyyzeuniqwi5wnhh3drm