EVALUATION OF DATA ASSOCIATION AND FUSION ALGORITHMS FOR\ud TRACKING IN THE PRESENCE OF MEASUREMENT LOSS

Tracking in multi sensor multi target (MSMT)\ud scenario is a complex problem due to the\ud uncertainties in the origin of observations. Solution\ud to this problem requires appropriate gating and\ud data association procedures to associate\ud measurements with targets. A PC MATLAB\ud program based on track-oriented approach is\ud evaluated which uses nearest neighbor Kalman filter\ud (NNKF) and probabilistic data association filter\ud (PDAF) for tracking multiple targets from data of\ud multiple sensors. For track-to-track fusion, state\ud vector fusion philosophy is employed. The tracking\ud performance in the presence of simulated track loss\ud and recovery as well as in clutter is evaluated.\ud During data loss PDAF performed better than\ud NNKF. In the presence of mild clutter and sparse\ud target scenarios, the NNKF and PDAF give similar\ud performance