CMU Robotics REPAIR program craft robot to fix bad pipelines

Focus on natural gas pipelines

The team at CMU's Biorobotics Lab, renowned for its previous work developing robots for various industrial applications, is now focusing on natural gas pipelines.

The project, funded by the U.S. Department of Energy through ARPA-E's REPAIR program, aims to create a robotic system capable of inspecting and repairing gas pipelines from the inside.

The motivation behind this endeavor is clear — the U.S. has a vast network of over two million miles of gas distribution lines, with repairs costing up to $10 million per mile.

The REPAIR program seeks to revolutionize the process by using robots and smart coatings to build new pipes within the existing ones, potentially saving millions.

The robotic system being developed by CMU stands out for its modularity and adaptability. Featuring a mobility module equipped with two-inch wheels and a third wheel on top, the robot can carry up to 60 pounds of payload, including the necessary repair materials.

While it may not be the fastest, inspecting nine miles of pipe in eight hours, its ability to make on-the-spot repairs at a rate of 1.8 miles in the same time frame is revolutionary.

What makes this system truly shine is the star of the show — a resin developed by the University of Illinois. Initially resembling soft-serve ice cream, the resin hardens within seconds of application to the pipe's inner wall.

CMU's robot is equipped with a spinning nozzle that applies the resin in a continuous bead, spiraling along the pipe as the robot inches forward. This innovative approach to repairs ensures efficiency and durability, meeting the DOE's requirement for repairs lasting at least 50 years.

Mapping, detailing, and identifying issues

The system also includes a mapping module with a high-resolution optical sensor that examines pipe color and uses lasers to make precise measurements, creating three-dimensional reconstructions.

Augmented reality aids engineers on-site in identifying defects and structural issues. Artificial intelligence processes the data gathered by the robot, utilizing simultaneous localization and mapping (SLAM) to produce highly detailed maps of the pipes.

The researchers have already tested their system using a 200-foot range, with the eventual goal of reaching an impressive two kilometers (approximately 6,500 feet).

Currently designed for 12-inch diameter pipes, a version for 6-inch pipes is in development, and the team envisions reducing the size to a challenging two-inch diameter in the near future.

As natural gas remains a crucial energy source, the CMU Robotics Institute's wheeled robot offers not just a glimpse but a tangible solution to the invisible challenges within our gas pipelines.

The potential for cost savings, efficiency, and longevity in pipeline maintenance is on the horizon, and as the press release by Carnegie Mellon University notes, researchers are leading the charge into a new era of infrastructure innovation.