An Analysis of Multiple-Object Tracking Using the Global Nearest Neighbor Method: A Case Study of Ground-Controlled Interception (GCI) Radar
Reynaldy Rizki Saputra, Friska Intan Muta'alliy, Adiyasa Nurfalah, Anies Marekan, Okky Aryoko, Maisevli Harika, Nurjannah Syakrani · 2023
GCI Radar plays a vital role in guiding interceptor aircraft to intercept unauthorized aircraft attempting to enter Indonesian airspace without permission. The operation of GCI Radar necessitates the implementation of an efficient multiple object tracking method to accurately determine the location of detected targets. The chosen multiple-object tracking method must exhibit a high level of accuracy and speed to effectively meet the demanding requirements of GCI Radar. The Global Nearest Neighbor (GNN) method has emerged as a prominent approach extensively employed in multiple-object tracking. Renowned for its computational efficiency, the GNN method proves particularly suitable for online tracking applications. The experiments conducted in this study demonstrate that the GNN method, with customized parameters, is capable of tracking high-speed objects within a speed range of 13.3 m/s to 531.9 m/s. It effectively performs track management and handles data with a time interval of 6 seconds. However, the method still created some false alarm and remained sensitive to noise. Overall, the findings of this research indicate that the GNN method is suitable for application as a multiple-object tracking technique in GCI Radar, as it aligns well with the characteristics of GCI Radar data.