Mixed Reality for Beyond Visual Line-of-Sight Bridge Inspection Using Robot-Assisted Nondestructive Evaluation (IM-5)

To improve data consistency, work efficiency, inspector safety, and cost effectiveness during routine inspections, drones have been increasingly used in recent years to support imaging and scanning over the surface of various elements in a bridge for surface condition assessment. Most drones are operated manually within a visual line of sight and unable to inspect river-crossing bridges completely since not all elements can be viewed by a drone operator using a binocular. Even for autonomous drones with collision-avoidance features, physical interaction with a bridge for nondestructive evaluation (NDE) is currently impossible in practice. An alternative solution with robot-assisted remote nondestructive tests in visually blocked areas would be desirable during detailed inspection and condition assessment of bridges. This project aims to develop a mixed reality (MR) interface that can streamline inspection process, analysis, and documentation for seamless data uses from inspection to maintenance in bridge asset management by automating access, visualization, comparison, and assessment, and to apply the MR interface in a beyond visual-line-of-sight (BVLOS) ultrasonic measurement for the thickness of steel girders from a climbing robot. Scope of Work in Year 1: (1) Develop a framework of MR-based bridge inspection for BVLOS elements (2) Integrate a NDE device into a climbing robot for remote bridge element inspection, and (3) Evaluate the MR-based inspection for thickness measurement of steel plates and lap-spliced joints to understand automated measurement precision, accuracy, and statistical variation.


  • English


Subject/Index Terms

Filing Info

  • Accession Number: 01847757
  • Record Type: Research project
  • Source Agency: Inspecting and Preserving Infrastructure through Robotic Exploration University Transportation Center
  • Contract Numbers: 69A3551747126
  • Files: UTC, RIP
  • Created Date: May 31 2022 5:10PM