Bridge inspectors may need to use several access methods and tools to adequately meet the objectives of a bridge inspection in accordance with governing National Bridge Inspection Standards (NBIS). However, some of these access methods, such as an under-bridge inspection truck (UBIT), can be costly to operate because the equipment is expensive to maintain and run and disruptive to traffic because it requires lane closures. Using an unmanned aircraft system (UAS) can be a cost-effective solution to obtaining stand-alone, high-quality visual inspection data, or to supplement standard inspection methods and equipment. Some UASs can also improve inspector safety and enable examination of areas that are difficult to access.
UASs can produce live streaming video, providing opportunity for the inspector to inspect while remaining on the ground. If inspectors identify deterioration in UAS images, they can then decide whether to perform a physical inspection to determine the severity and extent of the deterioration. Using UAS in this manner can provide more efficient use of standard access equipment and physical inspection techniques for assessing deterioration, in addition to increasing safety.
In an ongoing study, the Federal Highway Administration is conducting research to identify types of sensors used in UASs; quantity and quality level of data needed to perform satisfactory inspection using UASs; best practice guidelines for efficient and reliable use of the sensors; and guidance on how the collected data should be assessed, presented, and stored to provide reliable and actionable information to owners to support data-driven decisions. This research study also identifies the minimum requirements of sensors to provide comparable information as other visual inspection techniques.
“We felt it was very important to take a closer look at how State departments of transportation are using unmanned aircraft systems for bridge inspections because of the potential benefits of this technology,” says FHWA Executive Director Thomas Everett. “UASs are proving to be incredibly useful to bridge inspection staff to supplement inspection practices.”
FHWA expects to conclude the research project in March 2020. What follows are key findings of the research to date.
Components of a UAS
A UAS for bridge inspection includes the unmanned aircraft, control station, sensors, and pilot. A certified pilot is the most important piece of the system, controlling and flying the aircraft in a safe and professional manner. While not always a requirement, a visual observer can aid in scanning the sky to ensure safe flight while the pilot concentrates on the operation of the aircraft. As required by Federal law for all bridge inspections, an inspection team leader must be on site during the inspection.
 |
Optical cameras, infrared cameras, and LiDAR (light detection and ranging) systems are the most common types of sensors used. Depending on the tasks, an inspector can determine the appropriate types of UAS platform and sensor types. Optical sensors capture the imagery data (video as well as still images), which enable inspectors to see deficiencies in an up-close or magnified manner without having to physically access the specific area on the bridge. UAS-captured high-resolution images may reveal defects missed using distant visual inspection techniques. High-resolution imagery can also serve other purposes, from providing a record of surface defects to measuring and tracking some types of defects over time.
Infrared thermography (IR) sensors can detect areas of deterioration in concrete by identifying and viewing temperature gradients. Demonstrations have shown the areas of bridge deck delamination identified using IR sensors correspond well to the areas discovered using traditional sounding techniques.
Analyzing and Storing Data
When employing a UAS during bridge inspections, inspectors capture large amounts of data that require storage, post-processing, analysis, and dissemination. For most UASs, the imagery and data captured during a flight is stored on a removable media storage device, such as a secure digital (SD) memory card, integrated into the aircraft platform. The files stored on the SD card are a variety of common file types that are accessible by media-viewing and post-processing software.
 |
| MnDOT created this 3D bridge model with selectable image locations using data it collected with a UAS. |
Inspectors process the captured and stored data into different products to supplement inspection documentation, better inform decisionmakers regarding the structures, and improve future inspection planning. Common information products include images, video, 3D models, and surface models. Bridge engineers can use UAS imagery of the entire structure to create bridge “plans” for bridges that do not have records of the original structural drawings. Also, inspectors can use this visual information, and the associated geographic position information related to the images, to update the structure inspection records, identify and assess new deficiencies, track the extent of specific defects over several inspections, and update bridge repair recommendations.
In general, an inspector will use the standard inspection report format that complies with the NBIS, supports reporting data to the National Bridge Inventory, and satisfies State DOT policies and standards. When using a UAS to supplement an inspection, the inspector will select the imagery captured by the UAS sensor to include in the report. Thus, using a UAS for inspection purposes should not generate additional paperwork but the information and defects found in the images should be documented in the inspection notes and element condition data, as applicable.
“Data management can be the most challenging aspect of using a UAS,” says Joey Hartmann, director of the Office of Bridges and Structures with FHWA. “The substantial amount of data collected requires an appropriate data management plan to ensure the inspectors capturing the data have (1) a standard approach for collecting and transferring the data, (2) a known and secure location and structure for storing and retrieving the data, and (3) a well understood process for sharing the data and inspection products generated by the UAS.”
Cataloguing is the process of creating a directory of stored imagery files. It includes identifying where the data are located, identifying the types of data stored, establishing a process for version control, and instituting file naming conventions to which all users will adhere. A more advanced method of cataloging images is using a photogrammetric 3D model of the bridge, which requires creating a photogrammetric point cloud. This method is an alternative that enables all the inspection images for the bridge to be stored as a 3D model. Inspectors can select the bridge section of interest on the model (that is, where a defect exists) to view the image for analysis.
MnDOT tested this 3D modeling method to catalogue images. It enabled MnDOT inspectors to click on a point in the model and view images at that point to view defects. This can reduce the need for a manual photolog because the photogrammetry software will locate the image on the structure.
Future Advancements
As more bridge owners and inspectors incorporate UASs into their processes, the technologies available to improve inspections will continue to advance. For example, first-person view (FPV) devices or goggles are a relatively recent entry to the bridge inspection process. FPV gives the user a unique perspective from which to wirelessly view imagery and control the camera. Some FPV systems provide high-definition 1080p video and enable the user to control the sensor in real time with head movements. The image presented equates to looking at an 18-foot (5.5-meter) high-definition television from about 9 feet (3 meters) away. Some FPV systems also provide inspectors with the ability to digitally magnify the image, making it appear significantly closer and allowing a bridge inspector to see hairline cracks in the structure. For more information on FPV goggles for bridge inspectors, see “A New View for Bridge Inspectors” in the Summer 2018 issue of Public Roads.
Artificial intelligence (AI) is another technological advancement that inspectors may choose to incorporate into the UAS. AI can enable the system to navigate independently without human input throughout the structure (other than instructing the aircraft when and where it is supposed to fly and overriding the system in the event of a malfunction or signal loss). Flying the UAS in the same flight paths using AI can enhance the identification and tracking of defects over time. Inspectors could also use AI to collect and analyze many infrastructure images.
