Novel star identification method combining two star trackers with extended FOVs

This paper discusses a new Lost-in-Space star identification algorithm for multiple low-cost star trackers on small satellites. This new algorithm is inspired by the combination of two methods. One of them uses the gyro data to extend the images virtually, which has a defect in that it does not work well when the gyro errors are large. The other identifies the star pairs across multiple star trackers, which also has a defect in that it does not work well when the uncalibrated misalignment errors are large. By combining both of them properly, the new algorithm is expected to have robustness to both the gyro errors and the uncalibrated misalignment errors. A few new ideas are also incorporated in the combination method. By simulation, the feasibility of the algorithm is shown.