Adding Cabbie Know-How to Online Maps
Anyone who’s ridden in a taxi knows that cab drivers know their way around a city better than the average driver. They seem to have super-secret side-street maneuvers that shave minutes off a trip by avoiding traffic, lights, and other problems. Now researchers from Microsoft are mining cabbies’ knowledge to create faster driving paths for online maps.
For now it works only for Beijing, but a similar approach could work in many dense cities. The researchers analyzed GPS data of 33,000 Beijing taxis in hopes of finding faster driving routes that would even be practical for people who don’t drive at taxi speed or swerve recklessly between lanes. “These factors are very subtle and difficult to incorporate into existing routing engines,” says Yu Zheng, a researcher at Microsoft Research Asia. Zheng is an author on the paper describing the approach, called T-Drive, which is being presented this week at the International Conference on Advances in Geographic Information Systems, in San Jose, California.
Current drive-time predictions on online maps rely on the length of road and the posted speed limit. Some services will inform drivers that the route takes longer in traffic, but that doesn’t help someone who wants to know the fastest route from point A to point B, even if that route might look longer because it takes unexpected side streets. “This is the reality of all the Web maps,” says Zheng.
A handful of other projects have popped up to solve these sorts of problems. A collaboration between University of California and Nokia researchers collects GPS data from people’s cell phones while they drive to provide traffic information about side roads. MIT’s CarTel project in the Boston area incorporates data from drivers’ phones and from probes on taxis. And a startup in Silicon Valley called Waze lets people share their real-time driving paths with their online social networks, to help others choose faster routes based on current conditions.
Some Waze users are taxi drivers, but until now, no one has specifically targeted the group to mine their particular expertise. It’s an interesting approach, says Ehud Shabtai, cofounder of Waze. “You can probably assume routes taxi drivers take are more optimal, and it’s a good idea to learn more optimal routes by looking at them.”
The results seem convincing. According to the Microsoft researchers, the routes suggested by T-Drive are faster than 60 percent of the routes suggested by Google and Bing maps (which provide essentially the same driving time estimates as each other). Overall, T-Drive can shave about 16 percent off the time of a trip, the researchers say, which translates into about 5 minutes for every 30 minutes of driving.
The T-Drive doesn’t yet include real-time data such as accidents or new construction projects. This is something that the researchers say they can include in future versions of their system. Waze’s Shabtai also points out that a diversity of paths from ordinary drivers might be able to do a similar job. Either way you go, car trips might get shorter.
Geoffrey Hinton tells us why he’s now scared of the tech he helped build
“I have suddenly switched my views on whether these things are going to be more intelligent than us.”
Meet the people who use Notion to plan their whole lives
The workplace tool’s appeal extends far beyond organizing work projects. Many users find it’s just as useful for managing their free time.
Learning to code isn’t enough
Historically, learn-to-code efforts have provided opportunities for the few, but new efforts are aiming to be inclusive.
Deep learning pioneer Geoffrey Hinton has quit Google
Hinton will be speaking at EmTech Digital on Wednesday.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.