TRACIS: Real-time optical sensing & inspection for train track safety monitoring
Project Motivation
From interviews and visits to Singapore Mass Rapid Transit (SMRT) facilities, we have concluded that the existing method of manual track inspection has proven to be inefficient, inaccurate and is subject to human bias, error and fatigue. Especially with the planned expansion of transportation network, SMRT needs a cost-effective, portable and accurate solution that can be deployed to both opened and unopened lines. The solution to this problem will benefit public transportation enterprises around the world that suffer from similar maintenance problems.
Design
Our solution (TRACIS – Track Assessment and Crack Inspection System) can perform track scanning operations at the same quality compared to existing methods, but is more accurate (over 90%), portable and at a significantly cheaper cost (under S$3,000). TRACIS can empower SMRT to expand operations according to their needs and demands. Our system does not require any specialized training to operate as the track scanning function is done automatically by the onboard computer, so it can be easily added to be part of the standard track maintenance routine. By storing images and key information pertaining to the severity, location and type of cracks, we aim to create an AI model that can provide insights for predictive maintenance!
Awards
- Faculty of Engineering Innovation & Research Award (High Achievement) 2020
- Outstanding Undergraduate Researcher Prize 2020
Project Team
Students:
- Adhiraj Mahajan (Mechanical Engineering, Class of 2020)
- Cao Mai Ngoc Khang (Mechanical Engineering, Class of 2020)
- Chan Jia Jun (Mechanical Engineering, Class of 2020)
- Owyong Wei Liang, Shawn (Mechanical Engineering, Class of 2020)
Supervisor:
- Prof Hong Minghui (elehmh@nus.edu.sg)