Alejandro Pastor, GMV; Guillermo Escribano-Blázquez, Universidad Carlos III de Madrid; Manuel Sanjurjo-Rivo, Universidad Carlos III de Madrid; Diego Escobar, GMV
Keywords: fragmentation detection, track association, track correlation, cataloging, optical data
Abstract:
In June 29th, 1961, the Thor Able Star second stage exploded, two hours after separating from the Transit 4-A satellite. This event happened less than four years after the first launch of an artificial satellite, Sputnik 1, and generated a number of trackable objects three times greater than the number of known satellites so far in orbit. Since then, in-orbit fragmentations have been the most dominant source of space debris in the millimeter to decimeter size range. This is particularly critical for space safety, since objects in the upper part of this size range (centimeter to decimeter) are associated to the highest risk, given the current limits of shielding technology.
18th Space Control Squadron (SPCS) maintains one of the most complete and publicly available catalogs of resident space objects. It is published on SpaceTrack and contains more than 22,000 objects of which more than half are classified as fragmentation debris. The typical minimum size object that 18th SPCS can currently track is around 10 centimeters in low orbits and 1 meter at geostationary orbit, although these thresholds are expected to fall in the coming years due to the improvement of the Space Surveillance and Tracking (SST) sensor technology. The estimated number of objects between 1 and 10 cm size is not lower than 500,000 and thus we expect an important increase in complexity of catalog build-up and maintenance activities.
In this paper, a novel and operationally feasible methodology is proposed for fragmentation detection. A track-to-track association algorithm is used to solve the association problem arising from a fragmentation event. During the first stage after the event, the state of the fragments is expected to be very similar since not enough time has yet passed since the event for the fragments to separate. Fragments have not spread along the orbit, and therefore it might not be possible to distinguish individual objects, reason why they are usually referred to as debris cloud. Only once a sensor observation is outside the support of the expected object state, i.e.: the observable state separation of the fragments is greater than the one resulting from the expected sensor noise and dynamical model error, it is possible to start the identification, orbit determination (i.e., catalog build-up) and subsequent update (i.e., catalog maintenance) of the break-up fragments. Therefore, associations of tracks, or hypotheses, are generated, scored and pruned in real-time using several complexity reduction techniques. Fragments are identified, i.e.: hypotheses promoted, as enough sensing data is associated and the resulting orbit estimation is accurate enough to enable subsequent correlation of future tracks via track-to-orbit correlation methods. The use of this methodology during catalog maintenance operations would enable the automatic detection of fragments and reduce the detection time, which nowadays may take from weeks up to months.
Results are presented in optical survey scenarios with simulated data, providing clear association performance metrics and emphasizing on scalability and robust operational applications. The trade-off between detection time and orbit accuracy is discussed not only from the fragmentation detection point of view, but considering the whole catalog process, from the ingestion of new tracks to the update of orbital information. Besides, the sensitivity of key parameters such as track duration, observation geometry, parent objects orbit and fragments dynamics, among others, is assessed in order to provide recommendations for improving the detection of future explosion and collision events.
Date of Conference: September 14-17, 2021
Track: Astrodynamics