Feasibility study of extracting vibration characteristics in beams using Unmanned Aerial Vehicles (UAVs)

In recent years, there has been a growing interest into the monitoring of vibrations in bridges induced by moving vehicles or earthquakes, which results in deterioration and influences bridge deck cracking. The most widely used technique in structural health monitoring (SHM) for assessing bridge vibration characteristics is the use of accelerometers to measure natural frequencies and mode shapes. When inspecting long span bridges, the cost for this method is relatively high and requires specialized installation and data acquisition systems for acquiring the bridge data from various locations on the bridge deck.

This paper presents methods and results of an experimental study to investigate the feasibility of using unmanned aerial vehicles (UAVs), known as drones, to extract vibration characteristics in cantilever beams with different lengths, using computer vision methods. Using the drone to measure vibrations presents several challenges, including displacement stabilization and more specifically, the extraction of frequencies in beams. The vibration characteristics of the beams are captured using the camera mounted to the drone and the natural frequencies of the tested specimens are extracted using image processing algorithms. Accelerometer sensors are mounted to the beams, while a high-speed camera is also used to compare the frequencies. This study integrates sensors, data acquisition, UAVs and computer vision methods to investigate the potential use of unmanned aerial vehicles for measuring vibrations in bridge structures.