AeroLite-MDNet: Lightweight Multi-task Deviation Detection Network for UAV Landing

TL;DR

AeroLite-MDNet is a lightweight, vision-based deviation detection system for UAV landings that improves safety by accurately warning of landing deviations using a new dataset and evaluation metric.

Contribution

The paper introduces AeroLite-MDNet, a novel multi-task model with a multiscale fusion and segmentation for UAV landing deviation detection, along with UAVLandData dataset and AWD metric.

Findings

Achieves 98.6% deviation detection accuracy

Average Warning Delay of 0.7 seconds

Effective in real-world landing scenarios

Abstract

Unmanned aerial vehicles (UAVs) are increasingly employed in diverse applications such as land surveying, material transport, and environmental monitoring. Following missions like data collection or inspection, UAVs must land safely at docking stations for storage or recharging, which is an essential requirement for ensuring operational continuity. However, accurate landing remains challenging due to factors like GPS signal interference. To address this issue, we propose a deviation warning system for UAV landings, powered by a novel vision-based model called AeroLite-MDNet. This model integrates a multiscale fusion module for robust cross-scale object detection and incorporates a segmentation branch for efficient orientation estimation. We introduce a new evaluation metric, Average Warning Delay (AWD), to quantify the system's sensitivity to landing deviations. Furthermore, we contribute a new dataset, UAVLandData, which captures real-world landing deviation scenarios to support training and evaluation. Experimental results show that our system achieves an AWD of 0.7 seconds with a deviation detection accuracy of 98.6\%, demonstrating its effectiveness in enhancing UAV landing reliability. Code will be available at https://github.com/ITTTTTI/Maskyolo.git