Traffic sucks. Congestion is bad for the environment, health ( see: Beijing), and maybe most importantly, our sanity. One solution gaining in popularity amongst urban governments is the congestion charging scheme, in which drivers pay a toll to enter designated downtown areas. London, Stockholm, and Milan all have such systems in place, and New York came close to trying it out in 2007. Those efforts are all whippersnappers compared to Singapore’s program, which was introduced in 1975. Over the past few decades, the city-state has upgraded the system, which now uses radio transmitters to detect cars entering designated areas.
But two MIT researchers think they’ve got a better way of doing things. They’ve tested out a new method — and won an award doing it — that eliminates the need for the cameras and sensors in the streets that see where cars are going. That makes the system as a whole much more flexible and useful, since borders can be changed on the fly to reflect actual traffic conditions. As an added bonus, the system doesn’t just penalise drivers for entering certain zones, it helps them avoid them altogether.
The “RoadRunner” system, developed for Singapore by graduate student Jason Gao and his advisor Li-Shiuan Peh, issues a digital “token” to each car entering a congestion-prone area. Once a given number of tokens are assigned, a car can’t enter unless another vehicle leaves. Everyone else gets turn-by-turn directions to avoid the area. In computer simulations using data from Singapore’s Land Transit Authority, Gao and Peh saw an 8 percent increase in average car speed during periods of peak congestion. They also did a small scale test in Cambridge, Mass. to prove the technology works.
The exact details about how pricing would work aren’t important just yet. What’s innovative about RoadRunner is the way it liberates the tracking system from infrastructure. Singapore now requires all vehicles to have a dash-mounted transponder, which is read by radio transmitters on gantries (overhead structures like the ones that hold road signs and traffic lights) built at entry points to congestions zones. Gao and Peh fit cars with transponders roughly the size of a standard electronic-toll device like E-ZPass or FasTrak. They run on 802.11p, a standard similar to Wi-Fi but with a larger broadcast range, and communicate wirelessly with a central server. It may be possible to embed the technology directly into cell phones in the future.
The big advantage to this system is that urban planners can actively manipulate the location of congestion zones in real time, without the need for construction. “With our system, you can draw a polygon on the map and say, ‘I want this entire region to be controlled,'” Gao says. “You could do one thing for a month and test it out and then change it without having to dig up roads or rebuild gantries.”
That flexibility “has some very interesting implications,” says Sarah Kaufman, adjunct assistant professor of planning at New York University and digital manager at the NYU Rudin Center for Transportation. “It could help to re-route drivers around major temporary events (such as UN sessions here in NYC that gum up traffic quite a bit).” Tolling can be adjusted based on changing road priorities. The bit about the directions to avoid the demarcated zones is less impressive, Kaufman says. “The problem I see is that a lot of people are not driving through congested areas, but to them. Because congested areas are where their jobs are.”
Gao and Peh’s research landed them one of the best paper awards at the Intelligent Transportation Systems World Congress last week. French transportation expert Jean Bergounioux says RoadRunner is “worth getting into field trial as soon as possible to test and evaluate the feasibility of its industrial development and deployment.”