Asymmetric Cooperative Localization for UAV Inspection Inside Wind Turbine Towers Based on 3D Semantic Segmentation

View at publisher Summarise with VIRAT
Authors: Zhen Wang, Wei Deng, Rui Xia, Bo Xu, Bo Zhang, Runjie Shen
Cited by: 0
References: 10

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

To ensure the safety of wind energy infrastructure, an asymmetric cooperative localization framework is developed for UAV inspection inside wind turbine towers. Addressing the challenges of Global Navigation Satellite System denied environments and geometric repetitiveness, the methodology offloads perception and computation to a ground-based station. The approach involves a geometric preprocessing sequence to align point cloud coordinates and suppress platform pose drift, followed by a 3D semantic segmentation network to isolate the UAV from the complex tower environment. The UAV's position is then determined through geometric statistics of the segmented data. Experimental results in real-world scenarios demonstrate that the framework provides robust and stable position estimates, effectively distinguishing small targets in large-scale structured spaces. This research offers a practical solution for automated industrial inspections, reducing UAV payload while enhancing system reliability.

Fields of Study & Subjects

Robotics and Sensor-Based Localization Indoor and Outdoor Localization Technologies 3D Surveying and Cultural Heritage

Publication details

Journal / source
International Conference on Electrical Automation and Artificial Intelligence (ICEAAI)
Pages
653–658
Publication date
16 January 2026
Language
EN