||Mixed Reality-Assisted Element Level Inspection and Documentation
Bernard L. Kassner
Assessing the condition of VDOT’s inventory of 21,000 bridges and culverts requires substantial time and resources. Conventional assessment primarily involves periodic visual and tactile inspections conducted at “arm’s length” to quantify deficiencies that may require follow-up actions. However, there are emerging technologies that could potentially enhance the quality and efficiency of the bridge safety inspection, specifically, augmented and virtual reality (AR/VR) and artificial intelligence (AI). AR can provide historical data in a holographic image overlaid on top of the current image of a physical object, as well as access to element and defect codes prescribed in the AASHTO Manual for Bridge Evaluation. VR can enable remote inspections by recreating a visual, 3D model of a physical bridge in a virtual environment that can be remotely evaluated by engineers in an office setting. AI can provide automatic deterioration detection, measurement, and labeling of such deterioration as concrete cracks, delaminations, and spalls, and steel corrosion, in a way that is similar to human visual perception.
A system combining these technologies could automatically quantify defects and bridge element data for the NBIS inventory and generate the requisite inspection reports. Furthermore, this type of platform could accelerate the condition assessment process by aiding in measuring how much deterioration or damage has occurred since the last inspection, as well as provide a temporal, geo-located illustration of the changes. Such an evolution map could shed light on the mechanisms causing the damage and help to determine the required maintenance action. Additional benefits of the prototype will be the reduction of the impact of lane closures on the traveling public, enhanced collaboration amongst the inspection teams within and outside of VDOT, and enhanced training for future inspectors.