The U.S. Air Force collects information for the purpose of space surveillance through a global network of radars and optical sensors to maintain a catalog of over 20,000 resident space objects (RSOs), both active and inactive, with a minimum size of 10 cm. With the advent of better sensing technologies, such as the Space Surveillance Telescope and Space Based Space Surveillance satellite, much smaller RSOs can now be tracked than before. This will lead to the potential of at least an order of magnitude increase in the number of cataloged RSOs. A conjunction analysis is the process of quantifying the risk of close approaches between RSOs, which usually requires a probability of collision calculation. To perform this calculation the orbit states for every RSO with associated uncertainty must be maintained. Then comparative studies must be performed for two possible conjuncting RSOs. Several approaches, both heuristically-and statistically-based, have been developed to perform a conjunction analysis. However, with the rapid increase in the number of cataloged objects, traditional conjunction analysis approaches to determine the collision potential between all the tracked RSOs will not be computationally tractable, due to the computational complexity required to complete a full analysis of all possible conjuncting RSOs. Simple pruning methods, such as not comparing RSOs at vastly different altitudes, will not reduce the computational burden sufficiently. This paper will detail a new parallelizable method that reduces significantly the number of comparisons required in order to effectively and quickly identify RSOs with potential collision paths. By projecting RSO paths into a hierarchical discretized space, objects requiring a more in-depth conjunction analysis are suggested based on their probability of being located in the same grid space at some future time. Computation reduction, system architecture and scalability will be discussed. Simulation results that validate the proposed methodology are shown.
[1]
Salvatore Alfano.
Satellite Collision Probability Enhancements
,
2006
.
[2]
Robert Hall,et al.
Advances in Satellite Conjunction Analysis
,
2010
.
[3]
Russell P. Patera,et al.
Satellite Collision Probability for Nonlinear Relative Motion
,
2002
.
[4]
Hugh F. Durrant-Whyte,et al.
A new method for the nonlinear transformation of means and covariances in filters and estimators
,
2000,
IEEE Trans. Autom. Control..
[5]
Alan Genz,et al.
Numerical computation of rectangular bivariate and trivariate normal and t probabilities
,
2004,
Stat. Comput..
[6]
Salvatore Alfano.
Determining satellite close approaches, part 2
,
1994
.
[7]
Éloi Bossé,et al.
Robust combination rules for evidence theory
,
2009,
Inf. Fusion.
[8]
Salvatore Alfano.
Relating Position Uncertainty to Maximum Conjunction Probability
,
2003
.
[9]
Frank Bretz,et al.
Comparison of Methods for the Computation of Multivariate t Probabilities
,
2002
.
[10]
Felix R. Hoots,et al.
SPACETRACK REPORT NO. 3 Models for Propagation of
,
1988
.
[11]
BLRToN C. CouR-PALArs,et al.
Collision Frequency of Artificial Satellites : The Creation of a Debris Belt
,
2022
.