Ever felt that tingling sensation down your spine moments before you push the gas pedal down and zoom off even as the light turns red at a traffic intersection?
The next logical step after the immediate rush of adrenaline is a furtive look over your shoulders to check if any cop is on your tail.
That is if you haven’t already bumped into an oncoming car or an unfortunate pedestrian.
The latest is that now researchers at MIT know exactly when you are going to jump that red light!
The habit is deadlier than you think
Figures available with the National Highway Traffic Safety Administration say that back in 2008, 2.3 million accidents occurred at traffic intersections in the United States, which killed more than 7000 people. An estimated 700 of these deaths occurred as drivers jumped the red traffic lights. Unfortunately, half of them weren’t the red light runners, but innocent drivers and pedestrians. But now, a team of researchers from the Massachusetts Institute of Technology has come up with an algorithm that can predict when an oncoming car is most likely to run a red light. Looks like a lot of lives can now be saved.
Smart algorithm predicts your jump
A lot of things have gone into devising this algorithm. Factors like the car’s deceleration and distance from a traffic light helped the group identify potential cars that would possibly cross after the lights turned red at an intersection. The algorithm was also tested on data collected from a busy intersection in Virginia. It accurately identified the potential violators within a couple of seconds of them reaching a traffic signal. These precious seconds can be life-saving for other drivers at the intersection to react when alerted. The best part of this miracle algorithm is that it has a success rate of 85 percent, an improvement of almost 20 percent over all other similar algorithms.
Other things this could be used for
Professor of Aeronautics and Astronautics at MIT, Jonathan How, is enthusiastic at the idea that such an algorithm might be embedded in all future “smart” cars, making it easier for drivers to steer clear of violators. He, in fact, talks about a system that can warn drivers to stay put even when the light along their way is green since some other driver is trying to push the pedal. Thus, a collision is avoided. However, to make this a reality, cars have to first start “talking” to each other wirelessly. This wireless technology also known as vehicle-to-vehicle (V2V) communication is already being explored by the US Department of Transport along with major car manufacturers. Besides the V2V communication, such “smart” cars will also need short-range responders that continuously update location, speed, direction, and brake status.
A similar technology is the SignalGuru, which uses a smart phone camera mounted on the car dashboard to predict the best driving speed to keep up with the green lights along your route, besides displaying the gas prices as you pass stations.
And the next time you are at the wheel waiting for the lights to change, smile a little because the smart alecks who plan to jump are soon going to be extinct.