Georgia project uses 'roadbots' for highway repair
- By Kathleen Hickey
- Jun 28, 2012
Once the kinks are straightened out, the Georgia Department of Transportation may soon use robotic technology to automatically detect and seal highway cracks, improving worker safety and saving the state time and money.
The Georgia Tech Research Institute, which developed the prototype in conjunction with the agency, recently released a case study on the project, which began in 2003.
"Our prototype system has proved in many ways that a commercial-scale automated crack sealing system is viable," said Jonathan Holmes, the GTRI research engineer currently leading the project. "We demonstrated solutions to technical challenges — including the high-speed firing of nozzles, automated crack detection and navigation — in a real-time, limited-scale system."
Converted forestry app finds 'sweet spot' for highway repairs
With Boston app, potholes practically report themselves
Current crack-paving methods can be dangerous for workers, exposing them to traffic, and are slow. The new automated system requires only one operator and is faster than traditional methods, even though it only travels at three miles per hour.
It could also save highway departments money because sealing cracks maintains a road’s structural integrity, extending the time between major repaving projects, which are expensive, time-consuming and disruptive to motorists.
Highway departments have been using technology to try to improve the efficiency of their road repairs. In New Brunswick, Canada, the province’s transportation department converted analytics software used for forest management into a tool for taking a proactive approach to highway maintenance. Most highway departments fix roads at the end of their life cycles, a more expensive approach, said Kim Daley, New Brunswick DOT’s acting assistant deputy minister of corporate services and fleet management. By implementing transportation asset management technology, the department expects to save $1.4 billion over the next 20 years, she said.
The current GTRI prototype is limited to sealing thin cracks, which only require application of a liquid sealing compound, reported Gizmag. It also operates only on a narrow strip of highway, although it can be scaled up to handle an entire lane or road.
The trailer-mounted prototype uses an assembly, two different color LED lights and a stereo camera to create images that are then analyzed using algorithms to find cracks, even those smaller than one-eighth-inch wide. Based on the images, the operator can then continuously apply sealant via nozzles.
Before full-scale rollout, some issues need to be addressed, including improving the algorithms and imaging to increase the system’s accuracy in crack detection, which currently is at 83 percent, Holmes said.
"Our crack-detection algorithm was limited because we used a vision-based system, which was confounded by regions of high contrast caused by features other than pavement cracks, including dark stains in the pavement, lane stripes, raised-pavement markers, crack sealant and debris," he said. "A full-scale system may require a fusion of multiple imaging sensors, such as a 3-D laser scanning system."
The sealant dispersal system also needs to be modified, he said.
David Jared, acting chief of research & development at the Georgia DOT Office of Materials and Research, and GTRI principal research engineers Wayne Daley and Wiley Holcombe, research scientist Colin Usher, and research engineers Sergio Grullon and Steven Robertson also worked on this project. Georgia DOT’s report on the project is available here.