Stereo vision system and simulation for autonomous aerial refueling

The key issues of autonomous aerial refueling (AAR) based on stereo-vision navigation were studied. The important factors affecting the performance of AAR were analyzed, including the target's pose and the measure of relative distance. The stereo-vision was introduced into the visual navigation algorithm of the AAR, and the epipolar geometry principle was applied in order to accelerate the updating of the algorithm. The stability and capability of anti-disturbance were also improved. Furthermore, an integrated simulation system was designed. The simulation results show that the accuracy and stability of the vision navigation algorithm can meet the requirements of aerial refueling.