Jonathan Kos-Read/Flickr
City officials are starting to use cell phone towers to find out where people spend the most time idling in traffic, all via anonymous data collected from phones.
The data can then be used to clear traffic and give commuters back their sanity.
As part of a collaboration between AT&T Labs, the University of California at Berkeley, and the California Department of Transportation (Caltrans), researchers have launched a pilot program in the Bay Area with AT&T customers.
If it helps local city managers reduce traffic, the Connected Corridors Project project could expand nationally.
In California's Bay Area, sensors were first embedded in the road two decades ago to give commuters and the city relevant traffic information. Many of the sensors have started to fail.
Instead of repairing the sensors, which are bound to break again, AT&T and UC Berkeley realized cell phones would offer a more permanent solution. By tracking how close people's phones are to the nearest towers and recording how long they stay near those towers, analysts can draw a picture of where traffic is the worst.
"It's really only in the last couple of years that we're able to process all of this data and deliver it in a way that's effective and efficient for them," Chris Volinsky, assistant vice president of Inventive Science at AT&T Labs. tells Tech Insider. "So I think the need and the capability are coming together now in a nice way that we can provide solutions to the problems that are actually out there."
In Los Angeles, traffic managers are using the data to make an official traffic "playbook" - a standard guide for reading the city's traffic patterns in case there's ever an accident.
When one thing goes wrong, managers can turn to the playbook to see what other traffic systems might fail. Then they can decide which routes to open up.
The city can change the signaling on the on-ramps to roads, switch out the signage, and direct cars in a particular way. "They tell us that having our information to use to create these playbooks will help them navigate their daily bottlenecks and choke points in their networks to make their traffic routes more efficient," says Volinsky.
The Connected Corridors Project spans from northern California down to Los Angeles, and Volinsky hopes eventually to expand into New York City's subway system. Instead of relying on turnstile entrance and exit data, city managers will be able to see immediately where groups of people are going.
One big challenge: GPS data isn't available just yet.
Cell towers provide rough pictures of where people are - single points - not where they are going, an A-to-B. Without the use of satellites, the best traffic managers can do is guess the probability someone is taking a specific route by which towers they pass.
"Since we don't have GPS-level information, we have to make that inference of what roads people are on before we can provide useful information to people like Caltrans," Volinsky says.
The ultimate goal is twofold: to report to commuters when there are delays, in real-time, and to alert traffic managers when there are outages. If a train breaks down, travelers could receive a notification immediately. If there is an accident on the highway, the city can quickly change traffic light patterns to make another route more accessible.
And to assuage skeptics that claim grabbing personal data is a breach of privacy, Volinsky emphasizes that the bulk of the data is totally anonymous.
"Any AT&T customer has the right to opt out of data collection for any of these studies," he says. "The data we share externally is always done in aggregate: There were so many devices that went from this location to this location. So we like to think we're holding our customers' privacy in deep regard."