Autonomous Driving Systems Could Reduce Energy Savings and Environmental Benefits

Driverless vehicles could intensify car use — reducing or even eliminating promised energy savings and environmental benefits, a new study co-authored by a University of Washington engineer finds.

Development of autonomous driving systems has accelerated rapidly since the unveiling of Google’s driverless car in 2012, and energy efficiency due to improved traffic flow has been touted as one of the technology’s key advantages.

However, new research from the University of Leeds, the UW and Oak Ridge National Laboratory published in the journal Transportation Research Part A says the actual impact may be complicated by how the technology changes our relationship with our cars.

While vehicle automation undoubtedly offers some efficiency benefits, if people can work, relax and even hold meetings in their cars, they may drive more. That has the potential to erode the energy and environmental benefits of self-driving cars, the researchers found.

“There is a lot of hype around self-driving cars, much of it somewhat utopian in nature. But there are likely to be positives and negatives,” said co-author Don MacKenzie, a UW assistant professor of civil and environmental engineering. “By taking a clear-eyed view, we can design and implement policies to maximize the benefits and minimize the downsides of automated vehicles.”

The study analyzes self-driving technology combined with data on car and truck use, driver licenses and vehicle running costs to model the impact on energy demand of various levels of automation on U.S. roads by 2050.

The analysis identifies several efficiency benefits from self-driving cars and predicts ranges of likely energy impacts, depending on the extent of adoption of the technology and other factors:

• More efficient computer-directed driving styles (0 to 20 percent reduction in energy use)
• Improved traffic flow and reduced jams because of coordination between vehicles (0 to 4 percent reduction)
• “Platooning” of automated vehicles driving very close together to create aerodynamic savings (4 to 25 percent reduction)
• Reduced crash risks mean that cars can be lighter (5 to 23 percent reduction)
• Less emphasis from car buyers on high performance (5 to 23 percent reduction)

But the study also predicts that the very attractiveness of self-driving technology could reduce or even outweigh the efficiency gains.

It estimates a 5 to 60 percent increase in car energy consumption due to people choosing to use highly automated cars in situations where they would have previously taken alternative transport (e.g., trains or planes).

“When you make a decision about transport, you don’t just think about the out-of-pocket costs of the train ticket or the car’s petrol; you also take into account non-financial costs,” said lead author Zia Wadud, associate professor in the University of Leeds’ Faculty of Engineering and a research group leader in the University’s Institute for Transport Studies.

“Car owners might choose to travel by train to relatively distant business meetings because the train allows them to work and relax. The need to drive is part of the cost of choosing the car, just as standing on a cold platform is part of the cost of the train. If you can relax in your car as it safely drives itself to a meeting in another city, that changes the whole equation,” Wadud said

The study also predicts that people who currently find it difficult or impossible to drive, such as the elderly or some people with disabilities, will have increased access to road transport with the advent of the new systems, resulting in an estimated 2 to 10 percent increase in road energy use for personal travel.

Possible higher speed limits because of the improved safety of autonomous cars (7 to 22 percent) and demand for heavy extra equipment in driverless cars such as TV screens and computers (0 to 11 percent) might also tend to reduce efficiency savings.

A major uncertainty is the effect of autonomous driving technology on car-sharing. The technology could allow vehicles to move independently between different users and therefore not only increase sharing, but possibly also make it easier for users to match trip types to car types. Instead of using one car for all journeys, users might be able to use a shared, smaller car for a commute and a larger one for family leisure trips, for example. The authors say these factors could reduce energy consumption by 21 to 45 percent.

The study says many of the energy benefits of self-driving technology could be delivered by systems that still require the human driver to pay attention to the road and therefore do not radically alter transport decision-making.

The authors suggest that policymakers could focus less on accelerating the introduction of complete automation and more on promoting aspects of automation with positive environmental outcomes. For instance, regulators could encourage standardisation of car networking protocols to allow vehicles to communicate with each other on the road and therefore deliver benefits such as “platooning.”

The researchers warn that, if a high level of automation becomes the norm, it may be necessary to financially intervene in transport decisions. For example, self-driving cars’ navigation and communication systems could be used as a basis for road pricing schemes to control congestion and reduce overall travel demand.

“Vehicle automation presents a paradox: it may encourage people to travel much more, but at the same time it makes it practical to implement tools such as road pricing that can offset those effects,” MacKenzie said. “Ultimately, however, it’s up to government to set appropriate policies to manage these impacts.”

Co-authors include Paul Leiby, distinguished research scientist at Oak Ridge National Laboratory.

For more information, contact MacKenzie at [email protected] or 206-685-7198 FREE or Wadud through Montana Wright, Press Office, University of Leeds; +44 113 343 8373 or [email protected].

This article was adapted from a University of Leeds press release.