Sun Centroid Extraction Algorithm for Satellite based on Black Sun Effect

Abstract

Sun sensor is one of the most important components of attitude determination system for spacecrafts.An image of the sun is generated using a sun sensor based on a CMOS image sensor.To obtain the exact sun center, the image is analyzed using an image processing method; the sunspot coordinates are then given into a mathematical formula to produce the sunlight incident angle. As a result, the developed centroid algorithm plays an important role in the development of the sun sensor.Conventional digital sun sensor uses thresholding and filtering to extract sun vector. However, this method is susceptible to straylight and the albedo effect, which are two of the most common reasons of incorrect solar sensor readings. The thresholding method is not used here due to the difficulty of determining the best threshold for generating binary images. Instead, this study proposes that a black sun readout method be implemented onboard satellites where the black sun phenomena is caused by high illumination. The Black Sun method, which runs on the Raspberry Pi B+, is based on Template Matching, which uses the black spot as an indicator to locate the sun’s centroid. Despite the noise, the system extracts the sun centroid with more accuracy and reliability.