| Popis: |
Star sensors are highly reliable and represent the latest technology in the field of autonomous attitude determination. They are suitable for a broad range of missions, and the trackers are utilized in the commercial market of telecommunication geostationary satellites, earth observation satellites, scientific satellites and for interplanetary missions/probes. Although other methods for determining attitude exist, including sun sensors and magnetometers, they cannot report attitude with the precision a star tracker reports attitude. The present work aims to develop an algorithm for a star sensor using the overlapping grid method for star catalogue partition. The overlapping method is a modified sphere rectangle method which reduces the search through large number of grids. Since an actual sensor was not used, a star image was generated in terms of image plane coordinates. Previously developed algorithms have used an actual image from the sensor, and this present work can help optimize algorithms before implementation into the star sensor. A suitable methodological sequence was established, following which an algorithm was developed using MATLAB. A prefixed attitude was chosen, and the orientation angles based on the attitude were used to generate an image in terms of image plane coordinates. After processing it with the developed algorithm, the star coordinates from the ECI frame and sensor frame were input into the QUEST algorithm. The output of the QUEST algorithm is matched with the orientation angles from the prefixed attitude. This algorithm helps in reducing search and computation time, hence improving the response time of the satellite. |
