Authors:
Radu Danescu
1
;
Attila Fuzes
1
;
Razvan Itu
1
and
Vlad Turcu
2
Affiliations:
1
Computer Science Department, Technical University of Cluj-Napoca, Memorandumului 28, Cluj-Napoca, Romania
;
2
Astronomical Observatory Cluj-Napoca, Romanian Academy, Ciresilor 19, Cluj-Napoca, Romania
Keyword(s):
Space Surveillance, Real Time Target Identification, Low Earth Orbit.
Abstract:
Artificial objects in various orbits surround the Earth, and many of these objects can be found within the low Earth orbit region (LEO). This orbital zone also contains a significant amount of space debris, which pose a tangible threat to space operations, necessitating close monitoring. Various sensors can be used for either tracking, knowing the target’s orbital parameters and observing it for updating them, or for surveillance, which can also discover new targets. Real time identification of the satellite as it is detected by the surveillance systems provides a mechanism for selecting the targets for stare and chase applications, to decide if a new satellite has been discovered, or to identify a satellite that has outdated orbital elements. This paper describes a system capable of real time surveillance and satellite identification using limited computing power. The system relies on detecting trajectory endpoints at discrete time intervals, and then using these endpoints for frame
by frame trajectory prediction, which is then matched with detected tracklets. This way, the tracklets are identified in real time. The system has been tested by surveying real satellites, in real time, and the identification mechanism proved to work as expected.
(More)