Self-driving vehicles already improve traffic flow

We don't have to wait for 100-percent autonomous vehicles to see improvements in traffic flow.
Written by Kelly McSweeney, Contributor

New research shows that adding just a few self-driving cars to the streets can improve overall traffic patterns.

In the next several years, there will likely be an awkward transitional period when human drivers and autonomous vehicles share the roads. Consumer advocates, safety experts, and roboticists have expressed concern that mixing self-driving cars with human drivers could be dangerous. Recent field experiments in Tucson, Ariz., offer a surprisingly optimistic outlook. Autonomous vehicles help minimize stop-and-go traffic, which is a bad habit that we sloppy humans perpetuate.

"Our experiments show that with as few as five percent of vehicles being automated and carefully controlled, we can eliminate stop-and-go waves caused by human driving behavior," said Daniel B. Work, assistant professor at the University of Illinois at Urbana-Champaign, a lead researcher in the study.

Although there is an abundance of researchers figuring out how to build a self-driving car, nobody knows exactly how robotic cars will be introduced to public roads. Most industry stakeholders agree that, eventually, when all cars are fully autonomous, there will be major safety improvements. An often-cited statistic to back this claim is that 94 percent of fatal car accidents are caused by humans.

This may be true, but in the meantime, we need to learn more about how self-driving cars and human drivers will coexist. Robots thrive in controlled environments, but people are unpredictable at best and reckless at worst. There are some practical questions that remain unanswered: Will there be separate lanes for self-driving cars? Who will be at fault if there is an accident? How will car insurance work? Most importantly, is it safe to introduce self-driving cars (which are essentially gigantic robots) into an unstructured human environment?

The new study could help ease some of the anxiety about self-driving cars. During the field tests, one autonomous vehicle and more than 20 additional cars drove around a track. Typically, human drivers create stop-and-go traffic in this situation because of a phenomenon called a "phantom traffic jam." However, in this case, the autonomous vehicle drove at a steady pace, which eliminated waves of traffic. In the real world, this could help reduce overall fuel consumption, plus people would get to their destination faster.

Until all cars are autonomous, we will still have traffic jams caused by road construction, accidents, and rubbernecking. But the new research reveals that even by incrementally adding semi-autonomous features such as adaptive cruise control into today's cars, traffic can improve.

Related stories:

Editorial standards