Integrated probabilistic data association-finite resolution (Q1893020)

This paper discusses the problem of data association and track existence for sensors with finite resolution. An integrated algorithm called IPDA-FR (Integrated Probabilistic Data Association-Finite Resolution) is presented: It provides data association formulae and recursive formulae for probability of track existence. Its derivation follows the IPDA algorithm [the authors and \textit{S. Stankovic}, IEEE Trans. Autom. Control 39, 1237-1241 (1994; Zbl 0811.93057)] extending it to fit the finite resolution sensor model. They discuss two cases: the static clutter where a complete clutter map is available and the case where no a priori false alarm statistics are assumed. A simulation study illustrates the difference between IPDA-FR and IPDA algorithms in terms of true and false track discrimination and state estimation error.