Drogue pose estimation plays an important role in the docking process of aerial refueling. Many researchers have utilized vision-based methods, such as visible and infrared cameras. However, background signals like sunlight usually cause complex noise. Inspired by the fact that the atmosphere absorbs most Ultraviolet (UV) light around 254 nm, we propose a drogue pose estimation method based on a UV camera and asymmetric markers. Specifically, we place eight UV markers on the drogue and capture the scene image using a UV camera located at the root of the probe. To address the brightness variance of these markers, we propose a Region of Interest based adaptive threshold algorithm for UV images. By solving a Perspective-n-Points problem, our method can estimate the drogue pose. Simulated pose estimation and distance estimation experiments demonstrate that our methods are efficient and can run in real time. Overall, our approach offers an effective solution to drogue pose estimation in aerial refueling.
Drogue Pose Estimation Based on Ultraviolet Camera and Asymmetric Markers
Lect.Notes Computer
International Conference on Intelligent Robotics and Applications ; 2023 ; Hangzhou, China July 05, 2023 - July 07, 2023
21.10.2023
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Pose estimation , UV camera , Perspective-n-Point , Image threshold Computer Science , Artificial Intelligence , Software Engineering/Programming and Operating Systems , Computer Applications , Computer Communication Networks , Special Purpose and Application-Based Systems , User Interfaces and Human Computer Interaction
NTIS | 1974
|NTIS | 1974
|