Researchers: AI in connected cars relieves traffic jams at rush hour

As millions of people take to the highways this Thanksgiving, many will encounter traffic standstills for no apparent reason — no construction or accidents. Researchers say you’re the problem.

Human drivers just aren’t great at navigating heavy traffic conditions, but an artificial intelligence experiment in Nashville last week means help could be on the way. In the experiment, specially equipped cars were able to relieve rush-hour congestion on Interstate-24, researcher Daniel Work said Tuesday. In addition to reducing driver frustration, Work said less stop-and-go driving means fuel savings and therefore less pollution.

The professor of civil and environmental engineering at Vanderbilt University is among a group of engineers and mathematicians from US universities who have tackled the problem of phantom congestion after a simple experiment in Japan a dozen years ago showed how they did develop. The researchers there put about 20 human drivers on a circular path and asked them to drive at a constant speed. It didn’t take long for traffic to transition from a smooth flow to a series of stops and starts.

“Phantom jams are caused by drivers like you and me,” Work explained.

A person hits the brakes for some reason. The man behind takes a second to react and has to brake even harder. The next one has to brake even harder. The braking wave continues until many cars are stationary. Then, as traffic clears, drivers accelerate too quickly, resulting in more braking and another jam.

“We know that suddenly braking a car can have a big impact,” Work said.

Last week’s experiment showed that a few cars driving slowly and steadily can also have a positive impact.

The experiment used 100 cars running in loops on a 15-mile stretch of I-24 from about 6:00 a.m. to 9:45 a.m. each morning. Under the premise that if 5% of the cars on the road worked together they could reduce the prevalence of phantom traffic jams, the researchers outfitted these 100 cars to communicate wirelessly and send traffic information back and forth.

They also took advantage of adaptive cruise control, which is already optional on many new vehicles. With this technology, the driver can set a car to drive at a certain speed, but the car will automatically brake and accelerate when necessary to maintain a safe distance from the car in front. In the experiment, the adaptive cruise control was modified in such a way that it uses artificial intelligence to react to the entire flow of traffic, including events far ahead.

The car’s decision-making process happened at two levels, Work said. At the cloud level, information about traffic conditions was used to create an overall speed plan. This plan was then sent to the cars, which used artificial intelligence algorithms to determine the best course of action. Researchers were able to assess the impact of the connected cars on morning traffic flow using a dedicated 4-mile stretch of I-24 equipped with 300 pole-mounted sensors.

The experiment is a project of the CIRCLES consortium, a group including several automakers and the US Departments of Energy and Transportation. Other leading researchers work at the University of California, Berkeley; Temple University; and Rutgers University-Camden.

Liam Pedersen is Deputy Director General for Research at Nissan, a CIRCLES consortium partner who was in Nashville for the experiment last week. He said one of the exciting things about it is that it builds on technology that’s already in many new cars.

“This is not autonomous driving,” he said. “It’s something we could realize very soon.”

Asked if automakers will be willing to work together to ease traffic, Pedersen said, “I certainly hope so because the system works best when many, many cars participate.”

Last week’s experiment built on a work and colleagues conducted in 2017 at the University of Arizona. That repeated the Japanese experiment, this time with a single self-driving car thrown into the mix. The self-driving car smoothed the flow of traffic, requiring 98% less braking. This resulted in a 40% increase in fuel efficiency and a 14% increase in distance travelled.

The researchers are still calculating the numbers from last week’s experiment, but Work said it “showed that these congestion can be reduced by the novel automated vehicle technologies we’ve developed. There is no question that improved automotive technology can significantly reduce phantom congestion when implemented at scale.”

However, he warned that technology will not suddenly eliminate congestion.

“When there are more cars on the road than the road can carry, there will always be traffic,” he said. “But that can make those traffic jams less painful.”

© Copyright 2022 The Associated Press. All rights reserved. This material may not be published, broadcast, transcribed or redistributed without permission.

Leave a Reply

Your email address will not be published. Required fields are marked *