Skip to main content

New Research Uses Organic Computing to Make Traffic Greener

Putting advanced computing together with highly advanced -- and more widespread -- traffic sensors, researchers at Leibniz University in Hanover, Germany, have proposed an "organic" method of traffic control that responds to actual conditions. But is there an easier way?

Given that it's the day before what is (for us, at least) a long weekend, and I suspect that many of our U.S. readers are going to head out on the road someplace to celebrate the Fourth of July, it seems relevant to post a short item about everyone's favorite holiday byproduct: traffic.

If you were to take a poll of all of life's daily annoyances, I would hazard a guess that traffic would be way up there at the top (I personally would put overly long meetings at the very top, but then I don't own a car, so traffic is less of a concern for me...).

While there is no shortage of ideas in the works to try and reduce traffic jams, which in addition to being time-wasting and road-rage-inducing are also a big source of greenhouse gas emissions, a group of researchers at Leibniz University in Hanover, Germany, are proposing a kind of smart-computing, "organic" system that would allow traffic controls to respond to conditions in real-time to reduce traffic.

From an announcement posted on PhysOrg:
In the case of an urban traffic system, the sensors would be closed-circuit TV cameras mounted on road gantries and other places while the controllers, or actuators, would be traffic lights, which can effectively start and stop the flow of traffic.

Currently, traffic lights either have fixed timer controls or a centralised, control system. [...]

However, although these systems have been developed over many years, they do have several technical shortcomings and traffic jams do occur more frequently than drivers would like because of problems with flow control. Fixed timers are obviously flawed as they do not respond to traffic itself and even centralised systems cannot respond optimally to the changes in traffic movements out on the roads. This leads to jams and waste drivers' time, vehicle fuel, and to higher levels of localised pollution in towns and cities than might otherwise be present.
Researchers Holger Prothmann, Jurgen Branke, Hartmut Schmeck, Sven Tomforde, Fabian Rochner, Jorg Hahner and Christian Muller-Schloer have developed an organic computing model for a decentralized traffic-control system, one that in tests in Hamburg, has led to "significant" cuts in the number of times vehicles have to stop, in the length and frequency of delays and overall duration of car trips.

It's a cool idea, and certainly one that's much needed, especially in car-centric countries like the U.S. However, this research reminds me of a much simpler and even more intriguing idea I learned about earlier this year: cell phone-based traffic monitors.
The Mobile Millennium technology in action
mobile millennium
At CITRIS, a joint project between four branches of the University of California, a team of students have come up with a software program that turns your GPS-enabled cell phone into a traffic monitor. It's relatively cheap to develop, free to deploy, and...
The group found that, with just 2 percent of drivers on a given stretch of road using the software, the Mobile Millennium technology can out-predict the expensive (and often outdated) traffic-monitoring infrastructure put in place by transportation agencies.

Dan Work, one of the researchers on the project, explained that privacy is baked into the system: "We don't want to know how any given driver is driving, or where he's going," he said. The system just needs to know how the anonymous crowd as a whole is passing through an area.
While the report from the University of Leibniz is behind a paywall, you can download the Mobile Millennium software from http://traffic.berkeley.edu, and the site also lists which cell phones can run the software.

Happy driving, readers!

Photos CC-licensed by Flickr users fabrisalvetti and clicksense.

More on this topic