Red lights: we can't live with them, we can't live without them.
Or can we?
While the purpose of traffic lights - controlling the flow of traffic at intersections - is understandable, few would argue that having to stop at those pesky reds is irritating. They back up traffic. They make you late for work. And, as it turns out, they cause cars to consume more fuel and produce more emissions, between idling at lights and revving up again once they turn green.
Stopping at red lights could also soon be a thing of the past.
A group of researchers from MIT and Princeton University have developed an app that takes advantage of a growing trend: drivers who install brackets on their dashboards and mount their smartphones as GPS navigators.
The researchers used a network of these GPS-enabled cell phones to collect information about traffic lights. Based on images captured by the phones' cameras, the app is able to predict exactly how slowly a person needs to drive in order to miss the next red light.
The application, called SignalGuru, was tested on 20 cars in both Cambridge, Mass. and Singapore. The system used in Cambridge, where lights change according to fixed schedules, predicted the change of red lights to within two-thirds of a second. In Singapore, where traffic lights change depending on traffic flow, the system was less precise.
“The good news for the U.S.,” said Emmanouil Koukoumidis, the researcher who led the project, “is that most signals in the U.S. are dummy signals” — those with fixed schedules. But even a prediction to within two and a half seconds, Koukoumidis said, “could very well help you avoid stopping at an intersection.”
The test run in Cambridge was said to have saved drivers 20 percent more fuel. According to a statement from MIT, 28 percent of the energy consumed and 32 percent of carbon dioxide emitted in the U.S. comes from cars.
“If you can save even a small percentage of that, then you can have a large effect on the energy that the U.S. consumes,” Koukoumidis said. And as with all crowd-sourcing efforts, the system's effectiveness depends to some extent on the number of participating users whose phones can collect data.
The concept behind SignalGuru, leveraging information collected from users' smartphones, could be applied elsewhere, according to Koukoumidis. Cameras could capture location-specific information about gas prices, the arrival time of city buses, or about available parking spaces in a certain area.
While a commercially-available version of SignalGuru would tell drivers how much they should slow down to avoid stopping at a red light, don't expect an eventual app to tell users how fast they need to go to catch the next green light.
"We think that this application is not a safe thing to have," Koukoumidis said.
Photo: Flick/Magnus Akselvoll