A study by artificial intelligence startup Nauto surfaced this week explaining the perils of distracted driving and shedding light on the frequency of accidents occurring due to road distractions. The study also provides possible ways distracted driving can be countered through the use of technology.
To start, Nauto’s study claims that the numbers quoted by the National Highway Traffic Safety Administration (NHTSA) are underestimated. According to Nauto, NHTSA reports that there are an average of nine distracted driving casualties daily on U.S. roads.
“There are many studies on distracted driving, and the numbers vary quite widely. That is primarily because most of our traffic data here in the U.S., and also in other Western countries, is based on police-reportable collisions,” said Jennifer Haroon, the chief operating officer at Nauto. “For one, not all collisions are serious enough to be reported to the police. And two, when people are in a collision, even if distracted driving was one of the mitigating factors, they normally don’t volunteer and report it to the police.”
This makes the numbers quite inconclusive, though intuitively, most people understand distracted driving to be highly consequential. Concern related to the issues has led trucking fleets to create a variety of training programs for their drivers, emphasizing on the need to stay focused on the road while driving. This could also be evidenced through the National Safety Council’s initiative of announcing the month of April as the ‘distracted driving awareness’ month.
“Last year, we did a study with one of our insurance partners, and found that in a population of commercial fleet drivers using Nauto, that over 70 percent of collisions involved distracted driving,” said Haroon. “Our data pretty consistently shows that mobile phones are one of the primary reasons for distracted driving. Distraction can also be for various other reasons – like people doing paperwork in the cab, eating, drinking or even talking with others in the vehicle.”
Haroon explained that the first step to driving change within the industry was for fleet managers to acknowledge that distracted driving is a valid problem. She also advocated for the use of technology-based tools that give fleet managers video evidence and insights into driver behavior, which can be leveraged to help train drivers.
Though traditional telematics do help identify problems like hard braking, hard acceleration, idling, and even some forms of collision, they did little to help understand the issue of distracted driving. For example, a hard braking event could be due to two distinct incidents – for example, it could be because the driver was cut off by a bicycle and had to brake hard because that was the safest thing to do, or because the driver was distracted and happened to look up just in time to see traffic in front of the truck and thus slammed on the brakes.
“With traditional telematics, you will not get an answer to this. However, with the addition of cameras, you get added context on hard braking incidents. Computer vision and deep learning enable you to have real-time detection and provide real-time coaching to drivers on such events,” said Haroon.
Technology has improved substantially over the last decade, with artificial intelligence-based tools helping weed out dangerous driving practices and creating behavioral change among drivers. Haroon contended that Nauto’s in-cabin, real-time alerts helped decrease distraction by 40 percent on average, with even the bottom 10 percent witnessing considerable improvement.
“I think it is awesome that we can use technologies that we initially maybe thought were only going to be for automated vehicles, but we can use them today to help improve road safety for everyone,” said Haroon.