Statistical approach to track-to-track association problem

Abstract

We have applied techniques, developed previously (1997), to samples of track measurements and have obtained decision rule that are asymptotically close to decision rule characteristics (probability of false objects generation is not greater than any given significance level, and possibility to omit existing objects decreases with increase of observation time). The recursive computation algorithm that implements the above decision rule is developed, and objects location estimates are also obtained at every time step.