PAPER P03 · 2018
Autonomous UAVs for Structural Health Monitoring Using Deep Learning and an Ultrasonic Beacon System with Geo-Tagging
Kang, D., Cha, Y.-J. — Computer-Aided Civil and Infrastructure Engineering 33(10):885-902
Abstract read Citations: openalex: 386 · crossref: 369 · semanticScholar: 363 (2026-10-22)
Builds a UAV that navigates with ultrasonic beacons instead of GPS, so it can fly autonomously in places such as under bridges where GPS is unreliable, and pairs it with CNN crack detection. Detected damage is geo-tagged so its location on the structure is known. Concrete cracks were detected from video captured during autonomous flight.
Key findings
- Crack detection from autonomous-flight video: 97.7% specificity and 91.9% sensitivity
- Ultrasonic beacon positioning used as a GPS substitute in GPS-poor areas
Limitations
- Abstract only reviewed
- Beacons must be installed around the target, which limits use on large or inaccessible structures
- Validated on concrete cracks only