Aerial Contact Sensing for Improved Inspection of Transportation Infrastructure

Details

• Creators:
  Scherer, Sebastian
• Corporate Creators:
  Mobility21, Carnegie Mellon University
• Corporate Contributors:
  United States. Department of Transportation. University Transportation Centers (UTC) Program
• Subject/TRT Terms:
  Autonomous Vehicles Bridges Drones Human Factors Engineering Infrastructure Inspection Robot Vision Vehicle Design
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States
• Corporate Publisher:
  Mobility21, Carnegie Mellon University
• Abstract:
  The current state-of-the-art in using remotely controlled or small unmanned aerial systems (sUAS) for bridge inspection focuses only on taking visual measurements from a distance of many meters. This is not a sufficient substitute for the inspection made via contact sensing by a human inspector. Furthermore, these systems often require global positioning (GPS) or a prior map of the environment to operate autonomously. Many inspection jobs (e.g., underneath a bridge) may not satisfy these requirements. The goal of this project is to develop an autonomous sUAS capable of taking contact sensing measurements in GPS-denied and previously unknown environments. In this project the authors developed a first simple contact inspection system that has the potential to be low-cost and high-performance. This is achieved by using a novel configuration of tilted propellers with vision-based state estimation and control to allow flight without GPS. Additionally, the authors focus on intuitive human interaction with the system to select regions of interest (ROI) for contact inspection. Simple, useful feedback is given if the system detects something is wrong. Visual servicing control is used for the approach and retreat from the contact surface to minimize required prior knowledge of the surface and the environment. State estimation of the robot is performed by combining measurements from various onboard sensors, while avoiding the use of expensive and heavy equipment such as LiDAR.
• Format:
  PDF
• Funding:
  69A3551747111
• Collection(s):
  Autonomous Vehicles University Transportation Centers US Transportation Collection
• Main Document Checksum:
  urn:sha-512:2ccaa8c35a85c131504445e8a2d82a03a0b3a6ae7fb8e66b054c0e1a02ea45603450b03b09c73dcb1b588c2e0c1cf7ed05a7e65ee95f8cf9f6a21d0ecc1207b2
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/49465/dot_49465_DS1.pdf
• File Type:
  [PDF - 25.03 MB ]