Multi-Structure Mapping for Filtering Electric Arc Noise in Power Line Environments
Najlae Boulajoul, Alexis Lussier Desbiens, François Ferland
AI summary
Problem
Electric arc discharges near energized power lines generate short-lived LiDAR noise that accumulates as spurious obstacles, degrading drone navigation reliability. Existing environmental filters are either ineffective against unpredictable arc patterns or too computationally heavy for resource-limited aerial platforms.
Approach
The method leverages spatio-temporal consistency across consecutive LiDAR frames using a transient k-d tree to identify and suppress short-duration noise particles, while a persistent octree selectively integrates only enduring structural features into a global occupancy map.
Key results
- Up to 10x faster filtering than traditional outlier removal methods
- Mapping precision of 92.27% with F1-scores reaching 95%
- Consistent processing under 100 ms per iteration on embedded hardware
- Successful real-world flight validation near 315 kV energized conductors
Why it matters
Enables reliable, real-time LiDAR perception and safe autonomous navigation for drone-based power line inspection missions.
Abstract
Electric arc noise around energized power lines cor- rupts drone LiDAR measurements, accumulating in occupancy grids and producing spurious obstacles that degrade navigation reliability. Existing filters designed for environmental clutter such as snow, dust, and rain fail to consistently reject these short-lived arc transients and remain difficult to deploy on resource-limited platforms. We propose a dual-structure filtering framework that dynamically separates transient arc noise from persistent environ- mental features. Instead of filtering scan-by-scan, the proposed filter leverages spatio-temporal neighborhood consistency across consecutive LiDAR frames to suppress short-duration particles. A transient k-d tree accelerates neighborhood queries and removes arc noise around valid structures, while a persistent octree inte- grates only enduring features into the global map. Experiments show up to 10 times faster filtering and mapping precision of 92.27% with F1-scores up to 95%. Real-world inspection flights over energized power lines confirm that the approach maintains accurate, up-to-date maps and robust performance in the presence of electric arc noise.