Researchers from MIT and Princeton have developed a smartphone application called “SignalGuru” that uses the camera from a dashboard-mounted smartphone to capture images of traffic lights. Once the images are captured, they’re analyzed to detect whether the lights are green, yellow or red and then that data is passed along to other nearby SignalGuru users.
Using the resulting data, the app can relay to a particular driver how quickly he or she will need to drive in order to make the next light. If the next light is already red, the driver can coast up to it slowly instead.
(LIST: 50 Best iPhone Apps of 2011)
The researchers tested the app in Singapore, which uses dynamic traffic lights that change based on traffic levels and in Cambridge, Massachusetts, which uses dumb, old timed traffic lights spawned from the loins of none other than Satan himself (I live near Cambridge, believe it or not).
The results, according to MIT:
“By reducing the need to idle and accelerate from a standstill, the system saves gas: In tests conducted in Cambridge, Mass., it helped drivers cut fuel consumption by 20 percent.”
That’s pretty incredible, considering it requires no additional modifications to the car itself.
The app isn’t yet available to the public and when it is, it’ll have some challenges to overcome. For starters, it’ll face the chicken-and-egg problem most other crowdsourcing apps face: You need to have enough people using it to make the data reliable. If I’m the only person on Memorial Drive using the app, I’ll never know which upcoming lights are green or which of them are red. Actually, I’ll just assume they’re all red, since they’re ALWAYS RED.
The other obstacle is that the researchers can’t, in good conscience (and probably legally, too), release an app that tells people to speed up in order to make an upcoming light.
Still, the promise of cutting gas consumption by 20% may be enough to get a bunch of people to use this app, assuming it’s properly marketed. It may not even need to be directly marketed all that aggressively to consumers anyway, as the researchers see the technology perhaps being integrated into existing GPS routing software instead.
And the technology’s potential doesn’t stop there: It could be extended to “capture information about prices at different gas stations, about the locations and rates of progress of city buses, or about the availability of parking spaces in urban areas, all of which could be useful to commuters.”