Hasta La Gadget!

Sep 29

The Internet of Cars: New RD for Mobile Traffic Sensors

Posted in: Cars, Ford, Microsoft, mit, Today's Chili, Transportation

When we talk about “the internet of things,” we usually begin with commercial and household applications — tracking inventory, or a lost remote. But one future of networked objects might be in public information and infrastructure: the internet of cars.

For four years, MIT’s CarTel project has been tracking the driving patterns of GPS-equipped taxis in metro Boston. The research team, led by computer scientists Hari Balakrishnan and Sam Madden, thinks we can stop spotting traffic jams after the fact with news helicopters or roadside sensors by equipping cars themselves with position sensors and wireless connections. They’ve developed a new software algorithm that optimizes information-sharing between multiple nodes on a network, when those nodes are on the move, drifting in and out of close contact with one another.

Equipping cars with position and network technology has several advantages over traditional traffic-tracking methods. It’s already here, in the form of on-board GPS systems and the RFID fobs city car-sharing programs use to track cars and give multiple drivers access to vehicles. It’s less expensive than helicopters, and less static than fixed roadside sensors. Finally, news organizations and planners can see traffic tie ups as or even before they happen, rather than after the fact.

There are potential privacy concerns. Why should I allow the Department of Transportation, my local news team, or any entity to track my movements? Collection of this information would have to be closely regulated, highly encrypted, and strictly anonymized — perhaps even initially restricted to public and publically licensed vehicles likes public transit, cabs, police/fire/rescue vehicles, or cars and trucks owned by local government. The whole point is that when it comes to plotting traffic patterns, tracking unique users simply doesn’t matter.

But the potential upsides are tremendous. Having better knowledge of actual traffic patterns could help urban planners improve their transportation infrastructure, from retiming traffic lights to restructuring bus routes. It could help first responders and ordinary drivers avoid potential tie-ups.

Researchers at Ford and Microsoft are sufficiently intrigued. They plan to test the MIT researchers’ algorithm and network design in future versions of Sync, the Redmond-designed, Detroit-implemented automotive communication and entertainment system.