If you drive past potholes and faded lane dividers on your morning commute to work, chances are you’ll continue to see such road blemishes until someone alerts the local department of transportation to the problem by filing a complaint. Utah-based startup Blyncsy wants to help governments be a bit more preemptive than that.

The movement and data intelligence company is launching an AI-powered technology, called Payver, that will use crowd-sourced video data to give transport agencies up-to-date information on which roads require maintenance and improvements. Blyncsy is offering this service to governments at a reduced cost and with no long-term commitment.

Utah’s DOT will be the first to pilot the program beginning June 1, deploying Payver in the Salt Lake County region, which covers more than 350 road miles. Blyncsy will be announcing other pilots in different states over the next few weeks. 

Governments are typically a bit slower to adopt new technologies, and while the U.S. DOT has spent more than $ 250 million in public and private funds for smart city and advanced transportation options, much of the progress tends to revolve around making public transit in cities greener and more efficient. Blyncsy founder and CEO Mark Pittman argues inefficient road maintenance is not only unsafe, but it also causes higher carbon emissions. 

“The inspiration for Payver came in 2017 when the UDOT executive director set a goal that it would be the first department in the country to have real-time situational awareness on our roadways, and we’ve been working on solving that problem for them,” Pittman told TechCrunch. “They want to know what’s happening and when it’s happening automatically so the public doesn’t have to be involved. So if there’s roadside debris or stop signs missing or paint lines that need to be fixed, how does the department know without the public having to call and complain or without an accident occurring?”

Blyncsy’s Payver technology works by collecting any kind of HD images and videos from a variety of sources, such as Nexar dash cameras, and analyzing the data sets with machine vision to deliver output to customers. The insights are available to transit agencies in a dashboard format, but Payver also integrates into the maintenance management software that determines a rank order of repair jobs.  

For the UDOT pilot, Payver will initially focus on monitoring paint lines, which is the basic requirement to support an autonomous environment, but may expand into potholes, construction barrels, knocked over signs and whatever else can go wrong from daily wear. UDOT has a budget of about $ 90,000 for this pilot, according to Rob Miles, UDOT’s director of traffic and safety.

“Right now, we do a lidar scan of our roadways every two years, so we’re always out there collecting data of some sort, but it’s not at a high enough fidelity that we can manage our striping off of it,” Miles told TechCrunch. “We’re still really managing striping off of public complaints. We’re hoping with a different data collection system that we can move from that complaint-based system to something that has less opinion and more measurable data behind it.”

Pittman said optimizing active mobility forms by helping predict repairs needed for pedestrian crossings or safe locations for bike lanes will be a priority for Payver as the technology advances, which is important for maintaining equity and inclusion in mobility. Payver can also help bridge the racial and socioeconomic gap in road conditions, says Pittman, helping DOTs deliver equitable services to the entire public.

“Pete Buttigieg recently talked about how transportation as it is has helped support systemic racism at times and pigeonhole communities because of the way we build roads often with lower income populations being closer to freeways,” he said. 

“The same thing is also true when you think about roadway maintenance. Lower-income populations are much less likely to complain and higher-income populations are a lot more likely to complain, but the impact of a pothole in a low-income population means a broken axle, which can determine the viability of that family surviving. That’s not true in a high-income environment.”

TechCrunch