Performance analysis of camera rotation estimation algorithms for UAS sense and avoid

Several camera rotation estimator algorithms are tested in simulations and on real flight videos in this paper. The aim of the investigation is to show the strengths and weaknesses of these algorithms in the aircraft attitude estimation task. The work is part of a research project where a low cost UAV is developed which can be integrated into the national airspace. Two main issues are addressed with these measurements, one is the sense-and-avoid capability of the aircraft and the other is sensor redundancy. Both parts can benefit from a good attitude estimate. Thus, it is important to use the appropriate algorithm for the camera rotation estimation. Simulation results show that many times even the simplest algorithm can perform at an acceptable level of precision for the sensor fusion.