الفهرس | Only 14 pages are availabe for public view |
Abstract In Multiple Target Tracking (MTT) system. data association and tracking filter are two basic parts of tracking objects. The data association is an important key to associate the track to the true target in noisy recei\’ed measurements by rotating sensor at regular time interval. According to ome issues as in a cluttered environment. the reported measurements may be originated from true targets or from false alarm or clutter. In addition. the targets may be closely spaced. move with manoeuning. not detected in successive scans or to move in large groups. As a result. the association between the tracked target and true candidate measurement is difficult. Assigning wrong measurement to track often results in lost track and track break. Moreover. in dense clutter density the resulting number of false tracks may overwhelm the available computational resources of MTT system. For this reason. many data association algorithms have been de\’eloped to be the most po \overfuI techniques for these issues but still there are disad\’antages in their restricting assumptions. complexity. the resulting performance. Four data a sociation techniques widely used in MTT have some issues especially when tracking in heavy cluttered environment. The first technique is the Probabilistic Data Association Filter (PDAF) has good performance in low distri buted clutter densi ty but decreases its performance as the c I utter density increase. The second technique is the Joint Probabilistic Data Association Filter (JPDAF) which is an extension of PDAF to be used in multi-target tracking instead of single target as in PDAF. This technique suffers from high computational intensive as the measurements and tracked targets increase in addition to the tracked targets being to fail in |