Self-Organizing Traffic Lights

Prof. Peter Furth has won a $230,000 award from the National Science Foundation to develop self-organizing traffic signals. His research team will develop methods to make traffic signals more responsive to traffic, and to coordinate with each other using peer-to-peer communication. Traditional methods of coordinating traffic signals based on fixed cycle lengths and offsets tend to penalize pedestrians and public transportation because they require long signal cycles and lack the flexibility needed to give buses a green light when they need it. Because self-organizing logic allows traffic signals to coordinate without the constraint of a rigid cycle and to rapidly "heal" from interruptions that give priority for transit, it should provide a framework that lessens traffic delay for all users, especially transit users and pedestrians.

Source: News @ Northeastern

Peter Furth, a pro­fessor of civil and envi­ron­mental engi­neering at North­eastern, recently received a $260,000 National Sci­ence Foun­da­tion grant to study self-​​organizing traffic lights, which com­mu­ni­cate with one another over peer-​​to-​​peer net­works. These net­works could create safer and more effi­cient traffic pat­terns than existing sys­tems that cur­rently either use sen­sors to detect traffic or rely on coor­di­nated and timed loops—which can be dif­fi­cult to pro­gram and make travel less effi­cient for public transit and pedestrians.

  • How do self-organizing traffic lights differ from standard signals?

You use actuated signals —traffic responsive signals— as a base, but equip them with methods of peer-to-peer communication so they can organically organize themselves. They might form a wave of green lights one cycle, but the next cycle that might not work out. By communicating, one intersection can tell the next intersection, “I just turned green. Expect my cars to arrive in a certain amount of time.” They can try to coordinate, but they’re not fixed to a clock.

  • How can they make travel more convenient for buses, which tend to travel at a slower rate and stop more frequently than regular automotive traffic?

This can make a big difference because buses don’t advance at the same average speed as regular traffic. Buses are constantly interrupting things, but with self-healing logic, the signals can start to heal themselves and get back into their regular routine once the bus has gone by.

When we do try to give priority to buses in America—and only a few cities do—we’re often very cautious about it because it messes up the traffic patterns for everyone else. But with this, it’s easy to give priority to transit because the traffic lights can get themselves back into sync.

  • What are the first steps toward implementing this kind of system in an urban area?

The first step is to, in an experimental environment, show that logic like this can deliver a significant improvement. If that happens, then the Federal Highway Administration or a state highway department could test it out and use it. The signal controllers—the machines that we have out on the streets to control the traffic signals—all have embedded computers capable of applying this logic. But right now, we’re just trying to show that this kind of system could work.

Related Faculty: Peter G. Furth

Related Departments:Civil & Environmental Engineering