TY - GEN
T1 - Accurate Positioning for Refueling Plug with Coarse-to-fine Registration and Pose Correction*
AU - Bian, Jinyue
AU - Wang, Xin
AU - Liao, Zhiyu
AU - Ma, Hongbin
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Automatic refueling robot can free people from heavy work. However, work on the pose estimation of the refueling plug is relatively sparse. To this end, a novel precise positioning method for 6-DoF (Six-Degrees of Freedom) pose of the refueling plug is introduced based on coarse-to-fine template matching and pose correction. The coarse registration matrix is constructed by extracting 3D keypoints at the edge of the point cloud and combining with the geometric characteristics of the refueling plug. On the basis of coarse registration, fine registration is performed by aligning the main direction of the refueling plug handle. Through template matching, the grab coordinate system is constructed on the handle of the refueling plug, and the pose is corrected by modifying the grab coordinate system. The developed algorithm avoids building a large template library and reduces computing requirements. In addition, the accuracy and robustness of the 6-DoF pose estimation are proved on the automatic refueling robot.
AB - Automatic refueling robot can free people from heavy work. However, work on the pose estimation of the refueling plug is relatively sparse. To this end, a novel precise positioning method for 6-DoF (Six-Degrees of Freedom) pose of the refueling plug is introduced based on coarse-to-fine template matching and pose correction. The coarse registration matrix is constructed by extracting 3D keypoints at the edge of the point cloud and combining with the geometric characteristics of the refueling plug. On the basis of coarse registration, fine registration is performed by aligning the main direction of the refueling plug handle. Through template matching, the grab coordinate system is constructed on the handle of the refueling plug, and the pose is corrected by modifying the grab coordinate system. The developed algorithm avoids building a large template library and reduces computing requirements. In addition, the accuracy and robustness of the 6-DoF pose estimation are proved on the automatic refueling robot.
KW - 6-DoF pose estimation
KW - coarse-to-fine template matching
KW - coordinate system transformation
KW - grab coordinate system
KW - pose correction
UR - http://www.scopus.com/inward/record.url?scp=85151137668&partnerID=8YFLogxK
U2 - 10.1109/CAC57257.2022.10055996
DO - 10.1109/CAC57257.2022.10055996
M3 - Conference contribution
AN - SCOPUS:85151137668
T3 - Proceedings - 2022 Chinese Automation Congress, CAC 2022
SP - 2433
EP - 2438
BT - Proceedings - 2022 Chinese Automation Congress, CAC 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 Chinese Automation Congress, CAC 2022
Y2 - 25 November 2022 through 27 November 2022
ER -