Optimization of star research algorithm for esmo star tracker

Abstract

This paper explains in detail the design and the development of a software research star algorithm, embedded on a star tracker, by the ISAE/SUPAERO team. This research algorithm is inspired by musical techniques. This work will be carried out as part of the ESMO (European Student Moon Orbiter) project by different teams of students and professors from ISAE/SUPAERO (Institut Supe ́rieur de l’Ae ́ronautique et de l’Espace). Till today, the system engineering studies have been completed and the work that will be presented will concern the algorithmic and the embedded software development. The physical architecture of the sensor relies on APS 750 developed by the CIMI laboratory of ISAE/SUPAERO. First, a star research algorithm based on the image acquired in lost-in-space mode (one of the star tracker opera- tional modes) will be presented; it is inspired by techniques of musical recognition with the help of the correlation of digital signature (hash) with those stored in databases. The musical recognition principle is based on finger- printing, i.e. the extraction of points of interest in the studied signal. In the musical context, the signal spectrogram is used to identify these points. Applying this technique in image processing domain requires an equivalent tool to spectrogram. Those points of interest create a hash and are used to efficiently search within the database pre- viously sorted in order to be compared. The main goals of this research algorithm are to minimise the number of steps in the computations in order to deliver information at a higher frequency and to increase the computation robustness against the different possible disturbances.