Researchers in the U.K. are developing software that would allow cities to identify damaged roadways earlier.
As many of us have no doubt recently been reminded, these post-winter weeks can wreak havoc on road surfaces. As the thaw comes, so too do the potholes. Often, drivers are left with the costs for repairs after running over particularly nasty potholes, while city authorities spend big money fixing them. But what if you could find potholes before they even form?
Researchers at Nottingham Trent University in the U.K. are working to develop new technology that would do just that: monitor the surface of roads for underlying damage that may develop into a pothole if left unattended.
Research fellow Dr. Senthan Mathavan and his colleagues have developed software for analyzing data collected by road surface scanners run by British engineering company Dynatest. The partnership, which includes academics from Brunel University and the National University of Sciences and Technology in Islamabad, Pakistan, aims to make better predictions from finer pieces of data relating to road quality.
The scanners don’t just detect cracks, but rather the symptoms of raveling, the term engineers use to describe the gradual breakdown of a road surface. “Raveling is slightly different to cracks and potholes. Raveling is sort of a precursor to these things,” says Mathavan. “Once raveling happens it leads to serious defects like potholes and cracks and such.”
Dynatest's 2D and 3D scanners are simple enough in concept. They attach to the front of a vehicle, in this case a van, and scan the road surface for signs of raveling.
The scanners then provide the researchers with two different views of the surface of the road. The 2D scanner is effectively a high-resolution camera, while the 3D scanner creates a three-dimensional map of the ground.
“The lighting arrangements make sure that you get a good contrast between cracks and the regular road,” explains Mathavan. “From contrast, you can distinguish cracks.”
The scanners also work equally as well at night as they do during the day. But detecting evidence of raveling is one thing, as Mathavan explains. Making sense of all the data collected is the key challenge of this project.
In test runs, the researchers found that their software detected raveling correctly in 900 images that were collected. The idea is that if these sorts of defects can be caught much earlier than before, repairs and planning can in theory become more efficient or even cheaper.
Municipal governments have tried all sorts of ways of staying on top of potholes. Pothole reporting apps and hotlines have proliferated, and larger cities now routinely mobilize blitz-style repair teams who work overtime to fix up springtime roads as quickly as possible.
The technology these researchers are developing will eventually be open-source and could be adopted by governments and road authorities as part of their maintenance plans.
“In our view all these road authorities and highway agencies and people like that are moving more towards maintenance-based data,” adds Mathavan. “With funding cuts, they cannot reconstruct or overhaul these networks. So what we are trying to do is, we are trying to maintain this road network as best as we can.”
The next steps for the researchers include benchmarking the technology and comparing it against other technologies used by civil engineers, as well as traditional repair assessments carried out by humans, to gauge how their new algorithms can be improved.
The initial research from this study was only recently published, but the project has already garnered interest from councils and authorities in the U.K., says Mathavan, with whom they will be talking more over the “coming months.”
Right now the hardware used on the survey vehicles by Dynatest is quite large and bulky and isn't yet practical for use on public roads on a regular basis. Advancements, however, could make the equipment smaller, more discreet, and easier to attach to existing municipal vehicles for frequent assessments of a city’s road surfaces.