Telematics Provider Quantifies How Vision, AI Help Safety


Safety & Security

A truck hauling a wheel loader
Lytx has new data on how vision technology and AI help identify driving risks.

Telematics provider Lytx has announced new data on how their machine vision and artificial intelligence (MV+AI)-powered technology, used to detect and alert drivers and fleets of distractions and risky behaviors in real time, has enabled fleets and drivers to identify and correct significantly more risk, both inside and outside of the vehicle.

AI product debuted in 2015

The company first released MV+AI solution in 2015 to help fleets identify risky driving behavior like unsafe following distance. The most recent enhancements, launched this past February, include inside-view triggers, like cell phone use, powered by MV+AI to help solve the epidemic of distracted driving, a problem that costs approximately $175 billion a year in the United States, according to the National Highway Traffic Safety Administration. 

"Lytx's advanced MV+AI solutions make it easier for fleets to get started with and benefit from AI," said Sathya Kabirdas, research director, Frost & Sullivan. "Lytx is at the forefront of machine vision- and artificial intelligence-based video telematics for commercial vehicle fleets. The company has consistently delivered innovative technology in advance of the market and moved the space from speculating about the potential of MV+AI to deploying it at a scale that is helping transform safety and productivity for fleets of all sizes, across all industry segments."

Enhanced views for safety

Since launching its inside view triggers in February 2020, over 1,500 more fleets have deployed the company's MV+AI-backed driver safety solution, actively and accurately detecting high-risk behaviors among their drivers in real-time, while moving toward a safer and more productive future that empowers drivers.

"While many providers claim to have AI, it is really important to ask the right questions about the quality and breadth of their offering because many providers only capture one or two specific behaviors and do so using unreliable or limited datasets," said Dave Riordan, Lytx EVP and GM, enterprise business.

"This can lead to a barrage of false positives that result in inaccurate data and lead to trust issues with drivers, while increasing the risk of missing the real thing," Riordan said. "We are pleased to see so many fleets adopting our MV+AI technology, enabling them to detect and reduce cell phone use and other high-risk driving behaviors before they result in a collision, or worse."

Lytx says its technology is validated and backed by the largest and fastest-growing driving database of its kind, which grows by more than 140,000 new driving events each day, further training and improving its algorithms. This data is incorporated into a system that is able to deliver greater than 95 percent accuracy across 60 behaviors.

Mitigating safety risks

Compared to other video telematics, in which event capture is triggered by accelerometers alone, Lytx MV+AI enables fleets and empowers drivers to identify and correct significantly more risk, both inside and outside of the vehicle, including:

    No seat belt—identified 397% more

    Unsafe following distance—identified 332% more

    Incomplete stop–identified 243% more

    Food or drink—identified 173% more

    Failure to stop—identified 188% more

    Handheld device—identified 133% more

Lytx calculates that 35 percent of all near collisions captured by its devices involve one or more of these behaviors. This underscores both their prevalence and the risk they present, and in turn, the power of proactively capturing these habits–enabling driver self-correction in real time and the potential for deeper coaching in the future.

Source: Lytx