If you’ve ever seen two groups of ants meet up with one another on intersecting paths, you’ll notice that they don’t crash into each other. Instead, the larger group instinctively takes the right-of-way, followed by the smaller group – the same thing applies to bees and termites. Inspired by this behavior, Carnegie Mellon University telecommunications researcher Ozan Tonguz wondered if the same thing could be applied to traffic flow.
In development since 2009, his Virtual Traffic Lights system is based around the idea that vehicles approaching an intersection would use a vehicle-to-vehicle wireless network to communicate with one another. A patented algorithm would assess the number and direction of travel of the vehicles, and determine which group of vehicles heading in the same direction was the largest.
Vehicles in the larger group would then get a green light (or equivalent) on their dashboard screen – letting their drivers know to proceed – while cars in the smaller, intersecting groups would get red lights. Once the larger group had got through, the next-largest group would receive green lights.
According to Tonguz, simulations have shown that use of the system could reduce commute times by 40 to 60 percent during rush hours. Over the next two years, he plans on developing algorithms that will account for the presence of cyclists and pedestrians, along with creating a large-scale simulator, and testing the technology on actual roads.
The project has received US$2 million in funding so far, and a spin-off company has been established to develop the technology.
A somewhat similar system is being developed by the Dresden University of Technology and ETH Zurich. It would see traffic lights being left over the roads, but they would be able to communicate with the vehicles and each other, coordinating their signals to optimize traffic flow.
Source: Virtual Traffic Lights paper (pdf) via New Scientist