3D Displacement Monitoring of Railroad Bridges Using Unmanned Aerial Vehicles (UAVs)

Abstract

Railroad network is the main freight transportation systems in the US, with 40 percent of the total freight traffic being transported by rail. More than 50% of the railroad bridges in North America are over 100 years old. Nowadays, train loadings on bridges are higher than what they were designed to be in the past. Railroad bridges have been capable to adapt to new loads and successfully ensure safe operations by proper maintenance and repair since their inception, including continuous monitoring to proactively prevent any disruption in the transportation. Railroad managers have indicated that measuring the transverse displacement of the bridges under train crossing events can be used to quantify their proper performance and to inform management or repair decisions with objective data. Traditionally, Linear variable differential transformer (LVDTs) and other types of sensors are used to obtain the displacement data of the train, but these methods need human forces for their installation. Additionally, some parts of the bridges are not accessible or easy to reach for these types of inspections. Thus, aerial and remote sensing have become more popular. However, current methods for using Unmanned Aerial Vehicles (UAVs) to monitoring railroad bridges rely on cameras or lasers which are either technologically complex or economically unaffordable at large scale. This paper explains the development of a new hybrid sensing method to obtain total dynamic displacement using UAV. Sensors collecting displacements include lasers, cameras, and dynamic 3D cloud data acquisition. Displacement of the bridge is collected with the use of a low-cost laser sensor on the UAS with a camera attached to it. Then, the new algorithm extracts the hovering of the UAS and correct the laser displacement to obtain the displacement of the bridge under load. The results are compared to LVDT displacement for validation. The paper also identifies the priorities of North American Railroad owners regarding how the proposed technology can be implemented for new cost-effective management of railroad bridges maintenance.