Countering Driver Drowsiness
For a brief stint I worked 12-hour night shift job at a factory, five or six days a week. The factory was a good hour to an hour-fifteen minutes from where I was living at the time, and so this meant lots of driving while really, really tired. I have heard it said driving while tired can be just as dangerous as driving while inebriated, and after my several close calls working that shift, I can believe it. I had several near misses with deer, other drivers, and more than one occurrence of tires meeting the gravel and nearly putting myself in a ditch or a field. Every day was a struggle, as I would want to shower and get to sleep as soon as I got home, so getting coffee was out of the question. Typically, I relied on cranking up the air-conditioning â€“ even if it was already cold out â€“ and blasting the most aggressive, loud music I could tolerate. Even then, I was looking at one to two close calls per week, if not more. Today, technology exists that can help drivers remain aware, but a lot of it is cumbersome and expensive. Even though I do not work that job anymore, I know this is an issue that countless people have to deal with every single day, and having a more affordable and easy to manage method of drowsiness detection would potentially save a lot of lives.
Researchers Hans Van Dongen and Pia Forsman of Washington State University, Spokane have recently patented a new technology for just this purpose. This technology is based on the movements of the steering wheel, which have been shown to be a lot more sporadic in drowsy drivers â€“ rather than the currently available alternative that relies on video-based tracking of lane markers. This new technology uses relatively inexpensive, easy-to-install parts that include a sensor that measures the position of the steering wheel, which would easily be included as part of a factory installation or a not-too-difficult aftermarket modification.
â€śVideo-based systems that use cameras to detect when a car is drifting out of its lane are cumbersome and expensive,â€ť says Professor Van Dongen. â€śThey don’t work well on snow-covered or curvy roads, in darkness, or when lane markers are faded or missing… Out invention provides an inexpensive and user-friendly technology that overcomes these limitations and can help catch fatigue earlier, well before accidents are likely to happen.
In developing this new technology, 29 volunteers were put on a simulated 10-day night shift that was meant to cause moderate levels of fatigue, as was assessed by their performance on a common alertness/response test called the PVT, or psychomotor vigilance task. During each night of their shift, each of the volunteers spent four 30-minute sessions on a high-fidelity driving simulator, which would capture data for 87 different metrics that are related to speed, acceleration, steering, lane position, and many other factors. What these tests showed as that the two factors that best predicted driver drowsiness were the variability in the movements of the steering wheel and variance in lane positions. Using this data on the steering wheel variability the researchers were able to predict variability in lane position early on, which makes it possible to detect driver drowsiness before the car even drifts out of its own lane.
According to Van Dongen, â€śWe wanted to find out whether there may be a better technique for measuring driver drowsiness before fatigue levels are critical and a crash is imminent. Our invention provides a solid basis for the development of an early detection system for moderate driver drowsiness. It could also be combined with existing systems to extend their functionality in detecting severe driver drowsiness.â€ť
Image Credit: Thinkstock