Improved Binocular Localization of Bolts in Substation Based on Clamp and Bolts Detection Using YOLOv11 for Robotic Maintaining.

Saved in:

Bibliographic Details
Title:	Improved Binocular Localization of Bolts in Substation Based on Clamp and Bolts Detection Using YOLOv11 for Robotic Maintaining.
Authors:	Wu, Xuxiang¹ 2505889912@qq.com, Wu, Xiaoliang¹ 996591156@qq.com, Zhou, Qihui¹ 2751921426@qq.com, Huang, Haoming¹ 2724756219@qq.com, Gao, Changqing² gcq@hnu.edu.cn, Fang, Qiu³ qfang@hnu.edu.cn, Zhang, Xiaogang⁴ zhangxg@hnu.edu.cn
Source:	Engineering Letters. Jul2026, Vol. 34 Issue 7, p2488-2497. 10p.
Subjects:	Binocular vision, Object recognition (Computer vision), Triangulation, Stereo image processing, Clamps (Engineering), Electric substations, Robotics
Abstract:	Accurate bolt localization is a prerequisite for robotic maintenance in substations. However, conventional binocular vision methods often rely on global stereo matching and are therefore vulnerable to mismatches in complex outdoor environments with similar textures, and specular reflections. To address this issue, this paper proposes an improved binocular localization framework for substation bolts. First, YOLOv11 is used to detect bolts in stereo images and clamps in the left image. Then, a clamp-bolt pairing strategy and a clamp-guided stereo matching scheme are introduced to establish robust bolt correspondences while constraining the matching search space. Finally, the coordinates of bolts are recovered from the matched feature points using binocular triangulation. Experiments conducted in real substation scenarios demonstrate that the proposed method achieves millimeter-level localization accuracy, reducing localization errors by 47.1% and 43.7% compared with clamp localization and depth-image-based localization, respectively. The proposed framework effectively alleviates mismatch-induced localization errors and provides a reliable spatial perception solution for robotic bolt-maintenance tasks in substations. [ABSTRACT FROM AUTHOR]
	Copyright of Engineering Letters is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Accurate bolt localization is a prerequisite for robotic maintenance in substations. However, conventional binocular vision methods often rely on global stereo matching and are therefore vulnerable to mismatches in complex outdoor environments with similar textures, and specular reflections. To address this issue, this paper proposes an improved binocular localization framework for substation bolts. First, YOLOv11 is used to detect bolts in stereo images and clamps in the left image. Then, a clamp-bolt pairing strategy and a clamp-guided stereo matching scheme are introduced to establish robust bolt correspondences while constraining the matching search space. Finally, the coordinates of bolts are recovered from the matched feature points using binocular triangulation. Experiments conducted in real substation scenarios demonstrate that the proposed method achieves millimeter-level localization accuracy, reducing localization errors by 47.1% and 43.7% compared with clamp localization and depth-image-based localization, respectively. The proposed framework effectively alleviates mismatch-induced localization errors and provides a reliable spatial perception solution for robotic bolt-maintenance tasks in substations. [ABSTRACT FROM AUTHOR]
ISSN:	1816093X