Associating optical measurements of geocentric objects with a Genetic Algorithm: application to experimental data

Abstract

Currently several thousands of objects are being tracked in the MEO and GEO regions through optical means. This research aims to provide a method that can treat the association and orbit determination problems simultaneously, and is able to efficiently process large data sets with minimal manual intervention. This problem is also known as the Multiple Target Tracking (MTT) problem. The complexity of the MTT problem is defined by its dimension S. The number S corresponds to the number of fences involved in t he problem. Each fence consists of a set of observations where each observation belongs to a different object. The S≥3 MTT problem is an NP-hard combinatorial optimization problem. In previous work an Elitist Genetic Algorithm (EGA) was proposed as a method to approximately solve this problem. It was shown that the EGA is able to consistently find a good approximate solution when applied to simulated data. In this work the algorithm is applied to observations taken by the ZimSMART telescope of the Zimmerwald observatory.