جستجوی مقالات مرتبط با کلیدواژه « Uniform sky » در نشریات گروه « مکانیک »

The size of database and minimum number of visible stars in the field of view of star sensor are two important, influential and contradictory parameters that should be considered in design of star sensor. In this regard, the purpose of this paper is to unify the database using the uniform distribution of points on the celestial sphere with the triangulation method. For this purpose, the choice of the suitable star catalog, minimum suitable magnitude and elimination of double stars are the other steps of the uniformity process that is carried out in this study. Thus, the results of the investigations showed that Delaunay's triangulation method is faster and more accurate than the geodesic grid. Also, by simulating and performing Monte Carlo tests to count the number of stars observed in the different FOVs of a typical sensor, it was found that Delaunay's triangulation leads to a significant reduction of the probability of viewing the high density of the catalog stars in the field of view, so that the probability of observation more than 25 stars in all possible FOVs has reached to zero. On the other hand, for observing 4 or more than 4 stars at a confidence level more than 95% in non-uniform catalog, the field of view needs to be at least 12.5 degrees, while in uniform database; this field is slightly increased to more than 13 degrees. In other words, uniformity has increased the minimum field of view needed to see the minimum number of required stars.

Processor memory capacity and update frequency are one of the main restricting constraints in star tracker design and development. In order to decrease the volume of the data required onboard, uniforming the star catalog which is eventually used as pattern recognition database is considered. Three different methods of uniforming the star catalog have been applied. Spherical patches, fixed slope curve and charged particles or Thomson’s problem. After generation of a sphere with uniform distribution of points, a star is assigned to each point according to its spherical distance or best magnitude. In order to evaluate the performance of each method, seven evaluation criteria are defined. Point distribution minimum energy, catalog size, minimum star required for pattern recognition, mean and standard deviation of star distribution in each frame, database size and pattern recognition true recognize percentage. These seven criteria are combined in weighted equation of “average” to choose the best star catalog uniforming method with respect to the star tracker mission. After having implemented the average equation it is demonstrated that uniforming the star catalog using charged particle or Thomson’s problem has better results.