Toyota is developing a system called Guardian that uses a dashboard camera to check to see if a driver falls asleep. It’s part of a growing movement to increase automation in vehicles, but some experts say we’re a long way off from cars that are safe enough to fully drive themselves. “I’ve been a bit of a skeptic of full automation in terms of the timelines,” MIT professor John Leonard, who is working on Guardian, said at a recent MIT Mobility Forum, according to the news release. “[It] is going to take a lot longer to have this sort of ubiquitous robo taxi fleet, whereby, you know, a teenager today would never need a driver’s license or never need to have a real human Uber driver because all cars would drive themselves autonomously.”
Driving Minders
During the recent talk, Leonard showed off how the Guardian system can help keep drivers safe. It starts by recognizing the lack of driver awareness, takes over control of the vehicle, then, ultimately, reaches a point where—given an alert driver—the system no longer operates the vehicle itself. In another advance, Toyota researchers recently claimed to have successfully programmed a vehicle to autonomously drift around obstacles on a closed track. The idea behind this research is to utilize controlled, autonomous drifting to avoid accidents by navigating sudden obstacles or hazardous road conditions like black ice. “Our goal is to use advanced technologies that augment and amplify humans, not replace them,” said Avinash Balachandran, senior manager of Toyota’s Human Centric Driving Research in the news release. “Through this project, we are expanding the region in which a car is controllable, with the goal of giving regular drivers the instinctual reflexes of a professional race car driver to be able to handle the most challenging emergencies and keep people safer on the road.”
AI as Your Backseat Driver
Tal Krzypow, the vice president of product at Cipia, which uses AI and computer vision to monitor drivers for signs of distraction and drowsiness, said in an email interview that around 80 percent of all accidents are attributed to distracted driving. “We’ve all had experiences where we looked away from the road to grab our drink out of the cupholder, adjusted the radio, or were distracted by kids screaming in the backseat,” Krzypow said. “Humans can’t look everywhere at once, and our concentration is not perfect, so having AI systems to supplement our abilities as humans is not only important but life-saving.” Krzypow pointed out that in three seconds at 60 mph, a car travels almost 300 feet. He said that AI that can activate an emergency braking system to stop you from hitting the suddenly braking car in front of you could be the difference between life and death. Currently, most car AI systems have autonomous features to assist drivers and make the driving experience safer and more convenient. However, they’re not equipped to drive the car unassisted for long durations, Krzypow said. Examples of these systems include lane keep assist, emergency braking, Traffic Jam Assist, and Highway Driving Assist. Also becoming more common are Driver Monitoring Systems (DMS) that use AI and computer vision to monitor drivers for signs of distraction, drowsiness, and other dangerous situations, alerting drivers and regaining their attention to the road. Governments around the world are starting to mandate the presence of DMS. The EU has already passed legislation requiring DMS in new models beginning in 2025. The US Senate has introduced the SAFE Act, so this is no longer a “nice to have” feature and is rapidly becoming a mainstay in automotive safety, Krzypow said. Improved AI will help cars become more intelligent in the future, Siddhartha Bal, the director of autonomous mobility at iMerit, an autonomous car company, said in an email interview. “We will see much more focus on behavior analysis so that the car can judge the behavior of the people or any moving object around on the basis of their movements/intents,” Bal said. “That will make the drive even safer.”
