True 'smart car' learns as you drive

September 23, 9:25 AMScience News ExaminerMeg Marquardt

Previous Next

Comment Print Email RSS Subscribe

Subscribe

Get alerts when there is a new article from the Science News Examiner. Read Examiner.com's terms of use.

Email Address   Include other special offers from Examiner.com Terms of Use

Who's the smart one now? (Credit: Elijah van der Giessen)

When you grow tired, your driving habits change. Maybe your car lists more frequently into other lanes, or you take corners at more reckless speeds. Many times, you are unaware that you’re driving differently—making it all the more possible for accidents to happen.

A European coalition is trying to make those moments safer. They have built a literal smart car that learns as you drive and sounds warnings when your behavior changes. It is hoped that such a system, which can be adapted to any vehicle, will be able to alert drivers to the danger they may be putting themselves in while there is still time to act.

The system is contained on a microchip that is able to interpret incoming image data and create a 3D reconstruction of the situation. How a driver handles the situation—be it with sporty speed or conservative caution—is then recorded and stored.

Known as DRIVSCO, the project focuses on nighttime driving, when most accidents occur. As a person drives, the system learns how he approaches curves or avoids obstacles, cataloguing the driving style. At night, when a human’s eyes is no match for the car’s computerized night vision, the car will be able to alert the driver that they are not acting normally at a turn or when approaching an obstacle (such as another car).

The preemptive warning would give drivers time to react. If they still haven’t seen the obstacle, they could slow down their approach. If it is because they are growing tired and not focusing, the warning may be enough to shake them from a dangerous stupor.

The project is still in its infancy, but the initial tests have shown proof of concept: the smart-car-on-a-chip can understand the differences in driving habits of an individual. Professor Eduardo Ros Vidal, a scientist from University of Granada in Spain who works on the study, states, “We do not intend to develop automatic driving systems (as it would be very difficult for insurance agencies and car companies to come to an agreement in the event of a crash), but advanced driving assistance systems”. [Granada] The goal is simply to keep people safe—and alert—on the road.