Comparison-space selection to achieve efficient tracklet-to-object association. Issue 7 (1st October 2019)

• Record Type: Journal Article
• Authors: Siminski, J.A.; Flohrer, T.
• Is Part Of: Advances in space research. Volume 64:Issue 7(2019)
• Page Start: 1423
• Page End: 1431
• Publication Date: 2019-10-01
• Subjects: Tracklet association -- Cataloguing -- Curvilinear frame; Space sciences -- Periodicals; Astronautics -- Periodicals; Geophysics -- Periodicals; 500.505
• DOI: 10.1016/j.asr.2019.07.002
• Languages: English
• ISSNs: 0273-1177

Abstract

Abstract: A major challenge when maintaining a space object catalog is the proper association of new measurements to already cataloged objects. Optical observations are typically associated by comparing the modeled observation to the measured one. The modeled observation is generated from cataloged object states by propagating them to the epoch of observation and transforming them from state space, e.g. orbital elements, to the observation space, e.g. right ascension and declination angles. In addition to propagating the states, their propagated uncertainty distribution is transformed to observation space as well. Statistical distance metrics, such as the Mahalanobis distance, are then evaluated to test whether the observation originated from the cataloged object or not. These distance measures often assume that the uncertainty can be represented with a normal distribution. Assuming that the catalog state uncertainty is properly represented by a normal distribution, it can still loose this property during the propagation in time and the transformation to observation space. The uncertainty of the catalog state is typically much larger than the one from new measurements (only a few arc seconds for optical telescopes) and is therefore more affected by transformation distortions. It is therefore beneficial to perform the comparison in a space advantageous for the state representation. This study will present a projection-based transformation of tracklet information into a favorable frame around the cataloged object state. The effect of the comparison-space selection on cataloguing performance is assessed, i.e. it is systematically tested if it is beneficial to directly compare angles and angular rates, or to compare in the newly proposed projected frame.