Omnidirectional Solid-State mmWave Radar Perception for UAV Power Line Collision Avoidance

TL;DR

This work tackles the risk of UAV collisions with power lines by introducing a radar-only perception system that surrounds a small UAV with six mmWave radar modules to achieve omnidirectional sensing. It characterizes how such a multi-radar arrangement interacts with power lines and develops a robust, radar-tailored power-line avoidance algorithm, validated through outdoor field tests showing detection up to 10 m and successful avoidance at speeds up to 10 m/s, including wires as thin as 1.2 mm. The approach demonstrates a practical safety layer for both autonomous and manually piloted UAVs, enabling all-around detection and avoidance in challenging, weather-robust conditions. The combination of a lightweight, multi-sensor architecture, radar-specific behavior analysis, and an efficient avoidance strategy advances the deployment of safer UAV operations near power infrastructures.

Abstract

Detecting and estimating distances to power lines is a challenge for both human UAV pilots and autonomous systems, which increases the risk of unintended collisions. We present a mmWave radar-based perception system that provides spherical sensing coverage around a small UAV for robust power line detection and avoidance. The system integrates multiple compact solid-state mmWave radar modules to synthesize an omnidirectional field of view while remaining lightweight. We characterize the sensing behavior of this omnidirectional radar arrangement in power line environments and develop a robust detection-and-avoidance algorithm tailored to that behavior. Field experiments on real power lines demonstrate reliable detection at ranges up to 10 m, successful avoidance maneuvers at flight speeds upwards of 10 m/s, and detection of wires as thin as 1.2 mm in diameter. These results indicate the approach's suitability as an additional safety layer for both autonomous and manual UAV flight.

Figures (9)

• Figure 1: Top: Omnidirectional power line avoidance system flying near power lines with all six radar devices visible. Bottom: Time-lapse of power line avoidance maneuvers.
• Figure 2: Layout of the physical system from two perspectives.
• Figure 3: Visualization of the radar sensors' effective FoV and measurement errors relative to a 1 meter target in the $XZ$ (left), $YZ$ (middle), and $XY$ (right) planes. Dots represent measurements and shaded regions show the resulting effective FoV. Shades of regions with overlapping FoVs are a blend of the two overlapping shades.
• Figure 4: When pointed at an angle towards a power line, where does the radar sensor detect the power line?
• Figure 5: Plot showing the relationship of the angle between detected and closest point vs angle between boresight and closest point. While the radar boresight is still relatively close ($\pm$30°) to being perpendicular to the power line, the detected point is approximately equal to the closest point on the power line. As the boresight becomes even less perpendicular to the power line, the detected point moves further towards the boresight intersection with the power line.