(Technical Note) Design and Construction of Star Tracker Platform in Order to Satellite Attitude Determination Based on Star Tracking Algorithm

Attitude Determining is one of the major and critical satellites space missions. In this study, a new method to Attitude determination of satellites is presented. Such that, based on the proposed method search space will be more limited then accuracy and speed of attitude determination in the proposed method has risen. At first in this method, implementation and the test algorithms will be discussed, after these some algorithms, such as navigation, pattern recognition and ultimately attitude determination will be reviewed. In order to implement these algorithm. High quality images of stars which must provided by the star tracker camera requires to implement. Really these images to perform the necessary processing sent to the processor so the processor based on designed algorithms, determines the attitude of camera and satellite in all three axes. This means that some features considered for star tracker and based on them begins the designing process. The range of accurately determination for star tracker is one of these features. In this article, the ranges of two axes of Yao and Pitch less than 20 seconds on the scale of degree are considered and in the roll axis less than 100 seconds is intended. Can show in the results, much better accuracy and less than initial assumptions have been achieved. It also carried out by an adaptive identified algorithm so that the brighter stars are identified and based on their attitude determination, the sensor accuracy have increased. Because of according research, the clearer stars, have more accurate in calculation. The other important feature is the speed of attitude detection which performed by 1 GHz processor, and correct identification of pyramidal algorithm where have reached less than 15 milliseconds. Due to the duration, the desire update rate gained. Other important parameters which influence the accuracy of the attitude determination is knowing the exact coordinates of the intersection point vector of focal length lens with image sensors. By Land calibration for camera with a good accuracy, these parameters were estimated.