When we dial a number on our mobile at a speed of 120 km / h we travel, without seeing the road, a distance of almost 450 meters – a space equivalent to 4 football fields. If we write a message, we travel 660 m without paying attention to the road. Even if we do not perceive any risk, we are endangering our safety and that of others.
Now, new software is able to determine whether drivers are using their mobile while driving. The system, developed by engineering researchers at the University of Waterloo, in Canada, combines artificial intelligence (AI) with the use of cameras to accurately reveal when drivers are distracted performing other tasks, such as typing text messages, picking up something. from the rear seat, etc.
The algorithm that enables this software, based on previous research from the Center for Pattern Analysis and Machine Intelligence (CPAMI) at the University of Waterloo, is trained to recognize movement signals that alert the system to a possible risk. Frequent blinking, which is an indication of sleep, the position of the head and the movement of the hands are indicators that the driver may be taking reckless actions. The program evaluates these signs through various factors, such as their duration, and classifies them according to their degree of danger, determining which of them are a possible threat to the driver’s safety.
A system that will improve automated driving
“The new system can have a huge impact on society,” says Fakhri Karray, director of research and professor of electrical and computer engineering at Waterloo.
Data from the General Directorate of Traffic (DGT) show that in Spain at least 500 people die every year, 30% of the total number of victims, from traffic accidents caused by distraction while driving. And among the most common distractions is the use of the mobile. “This program could be incorporated into new automobiles with automated driving, so that it warns drivers of their recklessness and even takes control of the vehicle if danger is imminent,” explains Karray.
According to the AAA Foundation for Traffic Safety (American Automobile Association), when sending messages with the mobile, three forms of distraction are combined: visual distraction, cognitive distraction and manual distraction, actions that multiply the possibility of an accident by 23.
The current research of the center is now working on an improvement of the system, integrating all the functionalities in a single program and testing the use of more sensitive sensors that measure physiological changes, such as the size of the pupil and find variations in the heart rate to detect more precisely when the driver is clueless.
Those responsible for this project, Karray and his colleagues Safaa Bedawi, Arief Koesdwiady and Chaojie Ou, hope that this software will improve road safety in the future.
