Consideration on Actualizing the Non- Dimensional Star Pattern Recognition Algorithm for a Typical Laboratory Star Tracker

Star trackers are one of the most accurate attitude determination devices of a spacecraft. Star trackers are able to determine the spacecraft attitude with the use of recognizing the stars within their field of view. One of the major subsystems of star tracker software is the star pattern recognition algorithm. In this research a novel star pattern recognition algorithm called Non-Dimensional is considered. This algorithm can recognize the stars within its field of view without using the information of camera calibration. In order to fulfill this goal, the algorithm requires a database which contains the star inertial information. Afterwards a fast search method is used to compare the combinations on the image with database. Eventually there has been a tradeoff between database volume and the update frequency for a better performance.