Marker-Aided Multi-Sensor Fusion for Cable-Pulling Robot Localization in Substation Cable Trenches

View at publisher Summarise with VIRAT
Authors: Xiaofan Song, Haibo Huang, Jinfeng Zhang, Shun Li, Fei Guo, Congying Wu, Qi Li, Wenjie Xue, Xiaoxu Wang, Han Yao
Cited by: 0
References: 14

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Precise autonomous navigation in substation cable trenches is hindered by the GNSS-denied environment and the geometric degradation inherent to the “long corridor” effect, which limits conventional SLAM performance. To address these challenges, this paper proposes a reliability-aware multi-sensor fusion framework for tracked cable-pulling robots. Unlike standard integration schemes, an adaptive reliability gating mechanism is introduced within an Extended Kalman Filter (EKF) architecture. This mechanism dynamically filters unstable visual observations from discrete ArUco markers based on geometric quality, effectively mitigating errors caused by motion blur and occlusion. Field experiments in an operational substation demonstrate that the proposed method significantly outperforms inertial baselines, reducing the positioning Root Mean Square Error (RMSE) by over 64% and maintaining centimeter-level accuracy. The results validate that the proposed scheme effectively bounds localization drift, satisfying the stringent stability requirements for autonomous cable laying.

Fields of Study & Subjects

Robotics and Sensor-Based Localization Inertial Sensor and Navigation Indoor and Outdoor Localization Technologies

Publication details

Journal / source
International Conference on Electrical Automation and Artificial Intelligence (ICEAAI)
Pages
792–796
Publication date
16 January 2026
Language
EN