Traffic-Predicting Sensors Could Shorten Your Commute

Screenshot from new IBM software that could pave the way to shorter commutes. CREDIT: YouTube View full size image

A personalized, stress-free commute to work probably sounds like a dream to many metro travelers. But new technology that predicts future traffic could help pave the way to such an invention.

The technology, which was developed by IBM through a partnership with the California Department of Transportation (Caltrans) and the University of California, Berkeley, uses real-time data from sensors embedded in roadways to predict what the traffic is likely to look like in the near future. These predictions are then combined with a profile of an individual’s common travel routes to give personalized estimates of travel time and traffic conditions.

“It’s an analytical engine that ingests large amounts of sensor data and tracks historical trends … and then can read real-time data and match it to the history and say, ‘Oh, when we see this pattern this is what tends to happen 20 minutes later, [or] 40 minutes later,’” said John Day, manager of IBM’s Smarter Traveler program.

Television news and radio provide traffic reports, but the information is already old by the time it reaches viewers and listeners. “The thing that isn’t there today is something that tells you what it’s very likely to be in 30 to 45 minutes,” Day told TechNewsDaily. “That’s our key differentiator.”

For example, a popular traffic tool called Sigalert lets you log in and create personalized traffic reports for common routes. But just like other traffic tools, Sigalert gives data for what’s happening on the roads at that moment and not necessarily when you’re going to be on that particular road.

So far, the predictive traffic tool uses data collected in the San Francisco Bay area and is not yet available to the public. IBM says it is still working out the details of how to offer the service to drivers. The company also sees applications for the predictive software in other areas such as energy use. For example, the tool could be used to predict energy use (and potential bottlenecks) to help companies more efficiently deliver power.

This story was provided by TechNewsDaily, a sister site to LiveScience.