Phone app makes GPS look vague

Of course, GPS just isn’t accurate enough for the serially lost, as it only pinpoints your position to within ten meters.

To rectify this glaring shortcoming, Duke University computer engineers have developed a location-finding system that they say can work indoors and even tell if a user is on one side of an interior wall or another.

The system, dubbed SurroundSense, uses the phone’s built-in camera and microphone to record sound, light and colors, while the accelerometer records the movement patterns of the user. This information is sent to a server, which knits the disparate information together into a single fingerprint. As more people use the application, it gets smarter.

“You can’t tell much from any of the measurements individually, but when combined, the optical, acoustic and motion information creates a unique fingerprint of the space,” said Duke’s Ionut Constandache.

For example, in a bar, people spend little time moving, while the room is typically dark and noisy. In contrast, a Target store will be brightly lit with vibrant colors – especially red – with movement up and down aisles. SurroundSense can tell these differences.

The team fanned out across Durham, NC with their cellphones, collecting data in different types of businesses. “We went to 51 different stores and found that SurroundSense achieved an average accuracy of about 87 percent when all of the sensing capabilities were used,” Constandache said.

SurroundSense collects data at different times, so it can distinguish a Starbucks store at the morning rush from the slower period in mid-afternoon.

Currently, to collect data, the phone must be held with the camera facing down, though the researchers are working on this. And batteries are a bit of a problem, and the researchers are now considering the tradeoffs between having the application “on” all the time, which drains the battery faster, or having it take measurements at regular intervals.