Hybrid fuzzy probabilistic data association filter and joint probabilistic data association filter (Q1857067)

The paper studies a modification of traditional algorithms for target tracking in dense environments, namely problems with an additional uncertainty concerning the origin of the measurements besides the noise disturbance. The traditional algorithms of probabilistic data association filter (PDAF) used for single target tracking and joint probabilistic data association filter (JPDAF) used in the case of multiple target tracking are integrated with fuzzy \(c\)-means algorithms for the classification of a set of data into a given number of classes. The result is a combination of probabilistic and fuzzy approaches. The comparison of the proposed algorithms with the traditional ones is performed via simulation examples.