Design of a star tracker for nano/micro satellites

With increasing interest to the small satellites, the capabilities demanding from them also increasing. For several years now, the missions can be performed with about thousand kilogram satellites, can be accomplished with micro and nano satellites nowadays. Especially the number of nano satellites that send to space is increased 7 times in 2012. [5] Primary reason of this is the cost of the nano satellites are decreased to let them be developed in a university. Low cost of a small satellites lead up to many universities to develop their own Cubesat and stimulate the development of subsystems for nano satellites. Thus the capabilities of Cubesats begin to increase. Parallel to this advance, more accurate attitude determination and control required at Cubesats. To compensate this increasing attitude determination requirements, sun sensors, nadir sensors, magnetometers and gyros are used many times. However, the star trackers was not a part of this sensor set for a while. Because of their high cost, high mass and high power consumption rates it was not possible to use them on Cubesats. Nowadays it is available a few different start trackers systems can be used in nano satellites. Their cost is already high but with increasing availability of the components used in a star tracker they are becoming cheaper. In this thesis, with taking into account of this situation , a preliminary work has been performed to develop a star tracker with a low power consumption, low cost and low mass for Cubesats. Primarily a used friendly simulation environment developed to simulate the images taken with a star tracker. With using this simulation environment, images with different camera properties and various noise ratios are generated and projected to a screen. The algorithms that will be used are run on this images and the performance of the algorithms monitored. With the help of this simulation environment an efficient star detection algorithm is developed and a preliminary work for star tracking algorithm is performed. As Star tracker camera Raspberry Pi NOIR Camera V2 is used in tests. The reason of choosing this camera for test is its easy integration with Raspberry Pi development board. Although the cameras properties are different from a star tracker camera, it is considered sufficient for tests. The algorithms are developed and tested in Python language on Raspberry Pi board and on a desktop computer. Desired results have been taken in test environment and some improvements in algorithms foreseen as future work.