Study advocates "bilateral control" to disrupt traffic jams

A new study from the Massachusetts Institute of Technology suggests that new algorithms in vehicle technology could help alleviate traffic jams and improve traffic flow without making expensive improvements to infrastructure.

Researcher's from MIT's Computer Science and Artificial Intelligence Laboratory looked at the effect of "bilateral control" on traffic flow. In this type of traffic model, an equal amount of distance is left between vehicles.

Several previous studies have showed how tailgating can cause "phantom traffic jams," where congestion occurs for no discernable reason. When a driver does not leave enough following distance and hits their brakes, drivers behind them are also forced to slow down to avoid a collision. This causes a ripple effect in following vehicles until traffic comes to a standstill.

These studies have led to frequent recommendations that drivers leave enough following distance from the vehicle in front of them to maintain traffic flow. The MIT study looked at the effect of leaving adequate room from a following vehicle as well.

"We humans tend to view the world in terms of what's ahead of us, both literally and conceptually, so it might seem counterintuitive to look backwards," said Berthold Horn, an MIT professor who wrote the article on the study with postdoctoral associate Liang Wang for the journal IEEE Transactions on Intelligent Transportation Systems. "But driving like this could have a dramatic effect in reducing travel time and fuel consumption without having to build more roads or make other changes to infrastructure."

Researchers were inspired in part by the movements of starlings, which fly in dense flocks and coordinate movements without colliding. Horn suggests that this behavior results from starlings observing other birds all around them, not just in front of them.

The study says that one easy way to incorporate bilateral control into vehicles would be to update adaptive cruise control systems, which adjust a vehicle's speed to maintain speed while also keeping a safe following distance from the vehicle in front of them. Most systems only use sensors in the front bumper, but bilateral control could be programmed by adding rear sensors as well.

Horn says that even if a small percentage of vehicles incorporated bilateral control into their systems, traffic flow would improve considerably. The study suggests that drivers using this method could reach their destination twice as quickly as those focusing only on the vehicles in front of them.

Researchers also suggest that bilateral control would be an improvement over the "platooning" of electronically connected cars, which require detailed coordination as well as a large network of connected vehicles. Bilateral control would simply require new software as well as some minor hardware upgrades.

CSAIL plans to test the use of bilateral control in simulations to determine if this method results in improved traffic flow and safety. This research will be funded in part by Toyota.

Horn first proposed bilateral control in 2013, although he only applied this concept to individual cars. The more recent study looks at the potential effect of this method on several miles of highway travel.


Loading comments...
Hide Comments