TY - GEN
T1 - Tracking and position of drogue for autonomous aerial refueling
AU - Zhang, Jingyu
AU - Liu, Zhen
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/11/9
Y1 - 2018/11/9
N2 - In order to achieve autonomous aerial refueling, in this paper, we propose a method to solve the position and orientation of drogue based on stereo vision. The method can be mainly decomposed into three stages: initialization stage, tracking stage, and calculation stage. The pre-training model and the binocular constraint are used to initialize the position of drogue. Through the information fusion method, the position of drogue in each pair of images acquired by the binocular camera is located. After positioning the drogue, the center point of the ring region of the drogue is calculated by matching key points in binocular vision. The characteristics of this system is that it can operate simply and be easy to transplant which makes it apply widely. This system doesn't need to add cooperation logo on the drogue, which is convenient and feasible in actual projects. The method was carried out on a pair of ABB robots, and the experimental results demonstrate the effectiveness of proposed method.
AB - In order to achieve autonomous aerial refueling, in this paper, we propose a method to solve the position and orientation of drogue based on stereo vision. The method can be mainly decomposed into three stages: initialization stage, tracking stage, and calculation stage. The pre-training model and the binocular constraint are used to initialize the position of drogue. Through the information fusion method, the position of drogue in each pair of images acquired by the binocular camera is located. After positioning the drogue, the center point of the ring region of the drogue is calculated by matching key points in binocular vision. The characteristics of this system is that it can operate simply and be easy to transplant which makes it apply widely. This system doesn't need to add cooperation logo on the drogue, which is convenient and feasible in actual projects. The method was carried out on a pair of ABB robots, and the experimental results demonstrate the effectiveness of proposed method.
KW - autonomous aerial refueling
KW - binocular vision
KW - confidence judgement
KW - drogue tracking
UR - https://www.scopus.com/pages/publications/85058311443
U2 - 10.1109/OGC.2018.8529989
DO - 10.1109/OGC.2018.8529989
M3 - 会议稿件
AN - SCOPUS:85058311443
T3 - 2018 the 3rd Optoelectronics Global Conference, OGC 2018
SP - 171
EP - 175
BT - 2018 the 3rd Optoelectronics Global Conference, OGC 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 3rd Optoelectronics Global Conference, OGC 2018
Y2 - 4 September 2018 through 7 September 2018
ER -