Author

Zainudin Bonok

Subject

- Teknik

Abstract

In a multi-sensor radar system, some of the sensor measurements may be incorrect, due to the occurrence of false alarm and miss detection. So the number of targets and which the results of measurements derived from the target is not known for certain. The problem becomes how to determine the number of targets, which in the measurement results are mixed between the correct measurements (derived from the target) and the incorrect (false alarm). This problem is known as a multidimensional assignment problem. Solving this problem required a heuristic method. In this paper, the Ant System method is offered to solve the multidimensional assignment problem and the Nearest Neighbor Kalman Filter to estimate the target state. Three different target trajectories were used in the simulation using three identical radar. It is found that the Ant System algorithm is able to separate between true and false alarm. Furthermore, the process of estimating the target using the Nearest Neighbor Kalman Filter. Keywords: multi-target tracking, ant system, nearest neighbor -Kalman filter.

Publisher

ARPNJournal of engineering and applied sciences

Contributor

Ifan Wiranto and Zainudin Bonok

Publish

2018

Material Type

ARTIKEL

Right

baik