Artificial Intelligence Detects Distracted Driving Through In-Cab Monitors

Bison Transport is now testing new in-cab driver monitoring systems in order to detect distracted driving.

Seeing Machines Limited has earned a contract that will allow it to install its automated monitoring system into the fleet of one of Canada’s largest trucking companies.

Now, Bison Transport will have Seeing Machines’ Guardian technology implemented into 120 of its truck cabs. Bison employs around 2,500 truckers.

“Having Bison Transport as a customer affirms our view that fleets who have a culture of safety truly value Guardian as a way to improve that culture and keep their drivers safe,” said Dr. Mike Lenné, vice president of fleet and human factors for Seeing Machines.

As Bison has a “right to decide” policy for its drivers, the company originally tried equipping 14 of its trucks with the Guardian system in 2017, and is now expanding that testing to 120 trucks.

Garth Pitzel, Bison’s director of safety and driver development, said that in the first six weeks of testing, the fleet ran trucks with the monitoring on, but with vibration alerts and in-cab audio disabled. For the next six weeks, Bison had all the driver alerts activated.

After this period, the company saw a reduction of 67% in fatigued driving incidents, a reduction of 40% in distracted driving incidents, and a whopping 97% reduction in cellphone use (even though the company already has a hands-free cell phone policy in place).

Bison has been named North America’s safest fleet by the American Trucking Associations and the Truckload Carriers Association.

Seeing Machines’ driver monitor assesses a driver’s visual attention, their degree of drowsiness, and the possibility for risk with computer vision algorithms. To process a driver’s level of distraction or fatigue in real time, the monitor uses in-cabin sensors to track factors like eye movement and notify drivers with audio and vibration alerts.

If the in-cabin sensors detect any fatigue- or distraction-related event, analysts at the 24/7 Guardian monitoring center receive a video file. The footage is reviewed and analysts can immediately alert a fleet manager if one of these safety risks is confirmed.

Additionally, fleet managers can view flagged video clips at any time through the Guardian Live portal, but the system will only record video if it detects an issue.

Lenné and his team at Seeing Machines are hoping for more opportunities to further the company’s position throughout the continent.

“North America represents a massive opportunity for the Guardian technology, and the deal with Bison validates our revised approach with hardware pricing and service provision,” he said. “I look forward to seeing our team on this success as we rejuvenate our North American business and start to convert current conversations and interactions into Guardian contracts across this region.”

The Driveri system from technology supplier Netradyne detects distracted driving with object-detection methodology. The system analyzes a driver’s head movements in relation to body position, and when it measures what appears to be distracted driving, it notifies both the driver and fleet manager.

Adam Kahn, president of Netradyne’s fleet business, said Driveri analyzes everything from yawning to eye movement and gaze patterns in order to detect fatigue. Then, that information is processed in real-time on the device to allow for immediate communication with the driver.

Driveri also has a mobile app to let fleets easily address distracted driving with the driver.

Netradyne can also quickly identify both positive and risky behavior by using artificial intelligence, machine vision, and edge computing–a method that improves response time by bringing computation capacity closer to the location at which it is needed.

“Rather than selecting portions of the day and trying to determine the driver profile through happenstance, Driveri captures and analyzes every driving minute,” said Kahn.

However, the co-founder of Pronto.ai, a technology startup with similar monitoring technology, said these systems aren’t the overall solution to improving trucking safety.

“We cannot achieve the goal of safer roads by simply relying on driver monitoring technology alone to improve the effectiveness of existing technologies,” he said. “Instead, driver monitoring technology should be one part of a much broader culture of safety within a commercial fleet’s operations that involves rigorous coaching, training, and oversight by fleets of their drivers.”