Artificial system

Could artificial intelligence detect distracted drivers in Edmonton? U of A pilot program

A new pilot project in Edmonton is using artificial intelligence to detect drivers who use their cell phones while driving.

A three-week pilot project, led by the University of Alberta, aims to determine the prevalence of distracted driving in the city and test whether artificial intelligence is effective in detecting distracted drivers.

The City of Edmonton and the Edmonton Police Service are also participating in the pilot project.

The automated technology will be tested at a few locations in Edmonton, using sensors to capture high-resolution images through windshields.

The researchers said that in the pilot, anyone with a phone on them — whether in their hands, pocket or lap — would be considered a distracted driver.

“[The technology] look for a sign of the cellphone in the view of the windshield,” explained Karim El-Basyouny, research chair in urban road safety at the University of Alberta.

“If there’s a cell phone there and it’s moving, that’s kind of what it’s trying to detect.”

But El-Basyouny said it can be difficult for police to know how prevalent distracted driving is in Edmonton. He hopes the pilot can give researchers a better idea of ​​the extent of the distracted driving problem.

“At the moment the only way is to just use the human app,” said

“It’s not very easy for you to detect and…the police have to detect there’s a distraction and then they have to pull the vehicle to the side and then issue that ticket.”

According to data from the Traffic Injury Research Foundation, the number of fatalities caused by distracted driving exceeded those caused by impaired driving in Canada in 2019.

Distracted driving as deadly as impaired driving:

Australian company Acusensus developed the AI ​​the university is testing. Managing Director Tony Parrino said it has been used in Australia and several locations in the United States.

How it works?

So how does artificial intelligence catch drivers?

Parrino said his company shows the program examples of distracted driving, not distracted driving, and the technology uses an algorithm to learn.

The technology takes a photo of the potential incident on the road, which is then reviewed by a human. If the human decides the driver was distracted, that information would be forwarded to the police within 48 hours to issue a citation.

“With distracted driving, you can see the proof in the photo,” Parrino said.

“There’s no disputing it’s not like a speeding ticket where you have to figure out if the system is working correctly or not. It’s pretty obvious that the system is working correctly for distracted driving when you can see someone’s phone in his hand.”

At this time, the university is only testing the technology and those pictured will not receive a traffic violation.