Evaluating Storm Sewer Pipe Condition Using Autonomous Drone Technology
Abstract
The United States Air Force (USAF) owns a total of 30.9 million linear feet (LF) of storm sewer pipes valued at approximately $2.3B in its vast portfolio of built infrastructure. Current inventory records reveal that 78 percent of the inventory (24.1 million LF) is over 50 years old and will soon exceed its estimated service life. Additionally, the USAF depends on contract support while its business processes undervalue in-service evaluations from long-term funding plans. Ultimately, this disconnect negatively impacts infrastructure performance and overall strategic success, and the USAF risks making uninformed decisions in a fiscally constrained environment. This research presents a proof of concept effort to automate storm sewer evaluations for the USAF using unmanned ground vehicles and computer vision technology for autonomous defect detection. The results conceptually show that a low-cost autonomous system can be developed using commercial off the shelf (COTS) hardware and open-source software to quantify the condition of underground storm sewer pipes with an efficiency of 36 percent. While the results show that the prototype developed for this research is not sufficient for operational use, it does demonstrate that the USAF can leverage COTS systems in future AM strategies to improve asset visibility at a significantly lower cost.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2016
- Accession Number
- AD1054112
Entities
People
- Maria T. Meeks
Organizations
- Air Force Institute of Technology