Track-stitching using graphical models and message passing

In order to stitch tracks together, two tasks are required, namely tracking and track stitching. In this study track stitching is performed using a graphical model and message passing (belief propagation) approach. Tracks are modelled as nodes in a track graph trellis (lattice) structure. This graph is then solved by using a sequential Viterbi data association algorithm. A Kalman filter is used to perform tracking, as well as in gating operations and in determining the track-to-track association probability. Multiple crossing targets, with fragmented tracks, are simulated. It is then shown, that the algorithm successfully stitches track fragments together, even in the presence of false tracks, caused by noisy observations.