Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
SUMMARY Uncrewed Aerial Systems Applications for Bridge Inspections: Element-Level Bridge Data Collection NCHRP Project 12-122, âProposed AASHTO Guidelines for Applications of Unmanned Aircraft Systems Technologies for Element-Level Bridge Inspections,â sought to develop guidelines for state departments of transportation and other bridge owners regarding using uncrewed aircraft systems (UAS) to collect element-level inspection data. The objectives of this project were to (1) develop a selection process for UAS technologies, (2) develop operator and team qualifications, (3) develop a comparison between UAS and conventional element-level inspection methods, and (4) develop a stand-alone roadmap for holistic UAS implementation. The guidelines will provide organizational structure examples for a UAS program, airframe selection considerations, personnel training, bridge inspection method- ologies, data processing, and data storage. This final technical report includes a literature review, a review of multiple bridge inspections conducted with two UAS form factors, and a discussion of the major findings and conclusions. The literature review showed that UAS technologies have been widely used for inspec- tions of civil infrastructure, especially bridges, both in the United States and abroad. The most common systems are multi-rotor aircraft, which are well suited for bridge inspections because they can hover in place, are available in various sizes, and are equipped with a range of sensors. The most common sensors are visual cameras capable of recording video and still photos, much like those inspectors already use to document the condition states of bridge elements. Finally, the literature review identified knowledge gaps that were addressed in the inspection phase of the research, when two UAS form factors were evaluated to inform the selection process. The bridge inspections and the evaluation of the proposed methodology demonstrated that when applied in appropriate situations, UAS can improve the quality, safety, and efficiency of a Routine Inspection with element-level data collection. The two UAS form factors evaluated improved the safety and efficiency of an inspection when they could eliminate or reduce the need for other access methods such as aerial lifts, under bridge inspection vehicles, or rope- access techniques. UAS did not improve the efficiency of Routine Inspections that would typi- cally be conducted from the ground or bridge deck aided by binoculars, but UAS did improve data collection quality by providing enhanced access and better viewing angles. The data review process showed that most elements and deficiencies could be detected in videos and imagery collected by a UAS. As part of the inspection methodology, data was reviewed and cataloged in the field during the UAS inspection process. To verify data collection adequacy, UAS imagery and videos were shared with three qualified bridge inspectors for an in-office desktop review, during which the inspectors assigned quantities and condition states to the visible faces of major elements in the data. The quantities were only estimates, though, as the dimensions of the bridge elements were not shared with the 1
2ââ Uncrewed Aerial Systems Applications for Bridge Inspections: Element-Level Bridge Data Collection inspectors. The reviewers also noted that there were observed defects that warranted a closer inspection, confirming that a desktop review cannot be a primary means of inspection, and that data needs to be reviewed on site during the inspection in the event further investigation of a defect is needed. The data review process also showed that although artificial intelligenceâ based detection programs could assist bridge inspectors in the future, there is a need to develop specific bridge inspection models using data sets representative of the structural types, materials, and defects in the national bridge inventory.
