Open AccessEvaluating the Use of Unmanned Aerial Systems (UAS) to Perform Low-Slope Roof Inspections
Author(s) -
Dhaval Gajjar,
Joseph M. Burgett
Publication year - 2020
Publication title -
epic series in built environment
Language(s) - English
Resource type - Conference proceedings
ISSN - 2632-881X
DOI - 10.29007/jknh
Subject(s) - drone , roof , damages , photogrammetry , computer science , engineering , aeronautics , construction engineering , civil engineering , artificial intelligence , genetics , political science , law , biology
The use of Unmanned Aircraft Systems (UAS) or Drones are being explored for a wide range of civilian applications. The Federal Highway Administration (FHWA) recently reported that “construction inspectors that use UAS are reducing inspection time, improving effectiveness, increasing safety, and lowering costs.” If the FHWA is enjoying these benefits by leveraging this technology, it stands to reason that other industries that perform similar functions would also benefit. This study explores the opportunity of using commercially available UAS and structure-from-motion software to replace an in-person inspection for a low slope roof. The goal of the study was to see how much of a traditional in-person roof inspection could be replaced with a 3D photogrammetric model created from drone imagery. In this experiment, a SME roofing inspector identified deficiencies exclusively from a 3D model. Then, the SME inspected the low-slope roof in-person using traditional practices. The SME identified 191 specific deficiencies using 3D model and 200 deficiencies from the traditional method. The defects easiest to identify were open laps, alligatoring, punctures, wrinkles on the roof membrane and damages around the edges whereas total number and square feet of blisters, damages around penetration were the most difficult to identify in the model.