Viacheslav F. Fateev (Vympel Corporation), Sergey A. Sukhanov (Vympel Corporation), Yuri V. Burtsev (Space Forces), Zakhary N. Khutorovsky (Vympel Corporation), Sergey Yu. Kamensky (Vympel Corporation), Victor A. Stepaniants (Vympel Corporation), Alexander S. Samotokhin (Vympel Corporation), Vladimir M. Agapov (Vympel Corporation)
Keywords: Iridium, Cosmos Collision
Abstract:
The Russian Space Surveillance System maintains an independent satellite catalog and has extensive experience in solving all the types of problems related to the observations of satellites, in particular the problems of collision prediction. For dealing with this important problem the Russian SSC uses the direct method that has been presented at the First European Conference on Space Debris. This algorithm looks for the intervals of dangerous approaches of all the pairs of cataloged satellites and evaluates the geometrical characteristics of each approach and the probability of collision. The rather sophisticated formula used for the calculation of collision probability includes the sizes of approaching satellites, correlation matrices of position determination, relative positions and velocities for the time of minimum distance.
This work presents and analyses rather accurate approximate relationship for the collision probability, explicitly demonstrating the influence of all the parameters. The results of the analysis lead to the conclusion that only satisfactory coincidence of real and calculated errors of position determination for the satellites for the time of approach provides conditions for making the best possible decision on the possibility of collision. The paper describes the measures undertaken in the catalog maintenance software, which provides this coincidence for the satellites with insignificant atmospheric drag.
What potential capabilities for collision prediction do the Russian sensors and the Space Surveillance Center maintaining the catalog have? The collision that happened above the Russian territory on February 10 of the year 2009 at 16. 46 UTC of the non-cooperable satellite Cosmos-2251 and the operational spacecraft Iridium-33 provided data for the answer to this question. This paper presents the results. We performed the analysis of the time dependence of the errors of the determination of the relative positions of the satellites for the time of their collision for all the updated by radar measurements orbits, starting from 10 days before the collision. It is shown that for the whole time interval the theoretical values of the errors are in satisfactory coincidence with their real values. The average level of the errors in the direction of the radius vector for the time of the collision are about 20-30 meters and for other directions are less than 500 meters. For the 10 days time span before the collision the collision probabilities are calculated. They increase in general from 10-5 in the beginning of the interval up to 10-4 in the end just prior to collision.
It is shown that in case of collision warning starting two days prior to collision with further updates for each update of the involved orbits the frequency of false alarm will be about 0.02 (once in 50 days).
Date of Conference: September 1-4. 2009
Track: Iridium/Cosmos Collision