Tunnel Robot

• Zeyang Lei
• Jinrui Ma
• Xinnian Wang
• Ruiqing Yang
• Sirui Yang
• Chunming Yang

Advisor: Jonathan Komperda

Some underground tunnels are filled with pipes carrying steam and refrigerant. Navigating these tunnels to identify leaks or problem areas is inefficient and poses risks to workers. However, the development of robotics and computer technology allows for robots to do these inspections under extreme conditions, safely and efficiently. The goal of the project is to design an automated robot-type device that would work in the underground tunnel of the University of Illinois at Chicago to check for cracks and leaks in pipes carrying steam and refrigerant. NVIDIA’s jetson nano was chosen as the processor because of its strong information processing capabilities for real-time information processing and sophisticated Python development environments. An infrared camera, along with an ordinary camera, were used to detect leaks and eliminate the influence of other interference factors. The tunnel robot can follow a line in the tunnel, and synchronously record its position, while storing photos identifying problem areas, such as pipe leaks. Due to the harsh environment of the underground tunnel, the off-road performance, heat resistance, and moisture resistance of the electronic device were also taken into account. Accomplishing these goals will reduce the risks associated to working in underground pipeline tunnels and reduce costs by informing workers of necessary preventative maintenance.

See supporting documentation in the team’s Box drive.