A Measurement Weighting Method for Tracking Military Objects | ||||
| International Conference on Aerospace Sciences and Aviation Technology | ||||
| Article 38, Volume 13, AEROSPACE SCIENCES & AVIATION TECHNOLOGY, ASAT- 13, May 26 – 28, 2009, May 2009, Page 1-9 PDF (109.48 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/asat.2009.23535 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
| Ashraf M. Aziz; M. Moghazy; Ahmed M. ElBakly; M. H. AbdelAzeem | ||||
| Egyptian Armed Forces. | ||||
| Abstract | ||||
| Prior methods for tracking multiple military objects include various optimal and suboptimal two-dimensional assignment algorithms which make nearest-neighbour measurement-to-track association. This method works reasonably well in case of small number of targets and high signal to noise ratio. Another method is to assign a weight for each measurement and use a weighted centroid of those measurements to update the track. This method of weighting the measurements is known as all-neighbour and overcomes the disadvantages of the nearest-neighbour data association. Unfortunately, the computational complexity of an optimal all-neighbour data association technique limits its practical realization using even the fastest computers available. For this reason, many different tracking techniques have been developed which sacrifice optimal performance for the sake of computational feasibility. This paper proposes a new computationally feasible measurement weighting method to the problem of multiple targets tracking in a noisy environment. Computer simulation results indicate that the proposed weighting method successfully tracks multiple targets with a lower computational complexity and a little prior knowledge. | ||||
| Keywords | ||||
| Nearest-neighbor association; All-neighbor association; Kalman filter; Tracking; Measurement-to-tracks association | ||||
|
Statistics Article View: 121 PDF Download: 199 |
||||
