A Visual Marker Detection and Position Method for Autonomous Aerial Refueling of UAVs

Aerial refueling of UAV is an important project which can significantly improve the flight range and duration of UAV, meanwhile, the technology of relative position and attitude solution of tanker and receiver is the key technology of this project. Aiming at it, this paper carried out the research. Firstly, a concise circular visual marker is established according to the actual shape of the receptacle on the receiver. Secondly, this paper proposes a detection algorithm for the visual marker based on the state-of-the-art ellipse detection algorithm AAMED [1]. Thirdly, according to the structural characteristics of the visual marker and a fact, which is the rotation transformation does not change the eigenvalue of the coefficient matrix, the position and attitude information between the camera and the visual marker are solved, Finally, we conduct experiments based on the OptiTrack motion capture system. The experimental results verify the accuracy and practicability of the method proposed in this paper. Meanwhile, the time of processing one image frame is 56 ms on averagely, which meets the basic real-time requirement.