INVESTIGATION OF BRIDGE INSPECTION USING UNMANNED AERIAL VEHICLE

Abstract

Inspecting the bridge structure is one of the important steps in increasing the service life of the bridge. Visual inspection is the first step of the inspection that requires manpower to access the area to check for damages. As a duty of an engineer or an inspector, it has been required for them to have knowledge and Understanding bridge methods and behaviors to assess structural damages to bridges. Time-consuming, high-cost and individual have been taken into account for consideration, including the concealment of information. This research studies have presented Unmanned Aerial Vehicle (UAV) technology that has the
capability to reach and reduce accidents, as well as the ability to create 3D models. In this research, the principle of Photogrammetry has been presented to create a point cloud model (Point Cloud) to process Orthophotos that are effective in detecting bridge damages. The Ground Sample Distance (GSD) has been Configured resolution for 0.5 cm/pix, Overlap and Sidelap 80 and 80 are also respectively configured, including all 29 Ground Control Point (GCP) and Check Point (CP) assignments. The results have been validated for distances at each replicated point with 38 real areas, meanwhile damage patterns occur on the
model versus 26 actual locations. The results have shown another alternative outcome that could replace traditional audits. It increases opportunity to access information and reduce the risks that may cause harm to the speculators. Each point of the Orthophotos have had an average percentage discrepancy of 1.14%, and the damaged inspection have an average percentage discrepancy of 12.37%, compared to the actual damages on the bridge