University: Missouri University of Science and Technology (Missouri S&T)
Principal Investigator: Ruwen Qin, Missouri S&T
PI Contact Information: Phone: (573) 341-4493 | Email: email@example.com
Co-Principal Investigators: Genda Chen, Missouri S&T
Suzanna Long, Missouri S&T
Zhaozheng Yin, Missouri S&T
Sushil Louis, University of Nevada, Reno
Funding Source(s) and Amounts Provided:
Missouri S&T: $160,000
INSPIRE UTC: $160,000
Total Project Cost: $320,000
Match Agency ID or Contract Number:
Missouri S&T: In-Kind Match | INSPIRE UTC: 00061831
INSPIRE Grant Award Number: 69A3551747126
Start Date: January 1, 2018
End Date: September 30, 2020
Brief Description of Research Project:
Inspection and preservation of existing transportation infrastructure to extend their service life is an effective way of mitigating the pressure of steadily growing transportation demands on the aging infrastructure. Their current practice, though, represents one of the most costly operations in state departments of transportation.
The INSPIRE University Transportation Center will develop a remotely-controlled robotic platform that helps with these labor-intensive tasks and allows engineers to focus on decision-making processes. An important mission of INSPIRE is to leverage users’ capability of implementing, and interacting with, the robotic platform. Therefore, a long-term plan has been made to create a framework of training engineers and policy makers as well as new workforce on robotic operation and image analysis for the inspection and maintenance of transportation infrastructure. The proposed project, as a component of the plan, involves the prototyping of such a framework based on camera-based bridge inspection and robot-based maintenance.
Approach and Methodology. A training framework will be developed through an integration of the following components: (1) a set of training modules to use robotic systems for bridge inspection and maintenance; (2) data processing and pattern recognition algorithms in a semi-supervised way with engineers’ injection; (3) summary and visualization of processed inspection data, recognized patterns, and other research findings; and (4) a set of learning modules, which can be arranged as a customizable training plan for individuals, to help users analyze and use the vision-based materials in their decision making for bridge inspection and preservation.
Overall Objectives. This project aims to create a framework of training engineers and policy makers on robotic operation and image analysis for the inspection and preservation of transportation infrastructure. Specifically, it develops the method for collecting camera-based bridge inspection data and the algorithms for data processing and pattern recognitions; and creates tools for assisting and training users on visually analyzing the processed image data and recognized patterns for inspection and preservation decision-making.
Scope of Work in Year 1. (1) Collect bridge image data using cameras and annotate sample images; (2) Process image data and develop pattern recognition algorithms; (3) Summarize and visualize processed image data and discovered patterns; (4) Develop a training tool for image data analysis and understanding; and (5) Test and validate the prototype framework.
Scope of Work in Year 2: (1) semantic image segmentation to automatically detect bridge elements for inspection; (2) automatic retrieval of images containing the bridge elements that the inspector has selected and the analysis of these images for the inspector; (3) the use of transfer learning to update the semantic segmentation and image retrieval algorithms.
Describe Implementation of Research Outcomes:
Research outcomes and implementation plan will be described towards the end of this project.
Impacts/Benefits of Implementation:
Impact/Benefits of Implementation will be summarized at the end of this project.
Project Website: http://inspire-utc.mst.edu/researchprojects/wd-1/