It’s logical to think that highly connected and networked cars—talking to each other and the roadway—will naturally mean revolutionary improvements in vehicle efficiency and reduced emissions. Too bad that increased use of driverless cars may well cancel out those gains.
Assumptions of reduced environmental impact are based on the idea that computers, and digital networks, can do a much better job than humans at optimizing routes, sharing resources, and wringing out the best mileage possible. A lot of those efficiency improvements are likely to result from improved safety.
See also: Why Google's Driverless Car Is Evil
“If cars aren’t going to crash, then you can make them a lot lighter,” said John DeCicco, research professor at the University of Michigan Energy Institute. “You can get rid of all the things we’ve piled on to cars to make them safe.”
Don MacKenzie, an assistant professor of civil and environmental engineering at the University of Washington, agrees. “Improved crash avoidance means reduced crash worthiness requirements potential.” MacKenzie, who received his PhD in engineering systems from MIT and has researched fuel and powertrain efficiencies for more than a decade, estimates that removal of safety equipment and downsizing could mean efficiency gains of as much as 20 percent.
Roads Full of Super-Efficient Electric Mobility Bots
Vehicle automation also opens the possibility for greater deployment of smaller, less expensive special purpose vehicles—like the stereotypical futuristic mobility pods that usually come to mind. And because they are networked, many of these cars can easily be shared. Cars owned and operated by fleet companies—something like Uber with self-driving vehicles—but accessed on an as-needed basis, are also more likely to be configured for energy and cost-saving compared to today’s private cars souped up for fun and style.
DeCicco, a pioneer in rigorous evaluations of the environmental ratings of cars and trucks, believes that automated vehicles can increasingly be powered by electricity. That means even more gains in efficiency.
At the end of a drive, EV owners could simply send a car off for a few blocks, or miles, to find its own recharging stations. This allows people who live in multi-family dwellings to easily charge, whereas today electric car ownership and charging is easier for those with access to a garage. “It liberates the whole energy refill problem by totally automating it,” said DeCicco.
Then there is vehicle platooning, in which cars link up in train-like configurations for common legs of their respective journeys. MacKenzie pegs the benefit at as much as a 20 percent gain in efficiency, with perhaps another 20 percent benefit from eco-routing.
“Sometimes you see some big times from optimal simulation, said DeCicco. “But even those numbers are not a doubling of efficiency.” And of course, drivers would have to opt into driving in a slower and steadier fashion, rather than deciding to get to a destination faster.
Wait, Not So Fast
So far, so good for the efficiency of vehicle automation. “But there’s the other side of the story,” said DeCicco. “Automation could allow more driving.”
In other words, the ease of taking a ride in an automated vehicle could make us much more likely and frequently to jump in the car. “You take the driver out of the loop, and people will be much more willing to sit in their vehicles,” said MacKenzie. He suggested that the first 90 minutes at work is spent mostly answering emails. Why not decide to get a bigger, nicer and cheaper house 90 minutes away from your job, and answer your emails during the commute in your self-driving car? The result: more suburban sprawl and all the environmental damage it represents.
Increases in networked car sharing could add to vehicle traffic by those who might otherwise walk, bike or take public transit. “I don’t have to drive,” said DeCicco. “I can live even further away, and have my automated Barcalounger on wheels.” He envisioned a time when an exercise bike or rowing machine, as well as a big screen television, could become desirable vehicle features.
Matter of Human Priorities
MacKenzie, who has analyzed various automated driving pathways with colleagues from the U.S. Department of Energy’s Oak Ridge National Lab and U.K.’s University of Leeds, believes that all of these factors could lead to as much as 160% more travel—eclipsing the efficiency benefits of automated systems.
The increase in vehicle use (and related emissions) includes new trips by young, elderly and disabled travelers who previously lacked mobility. He also mentioned that vehicle efficiency on highways could go down, if speed limits are increased as high as 100 miles per hour, due to improved safety.
“If you’ve provided mobility to a new group of users, and you're saving lives, that’s great,” said MacKenzie. “But it’s premature to say we know how it’s going to affect energy consumption.” For MacKenzie, it’s two forces fighting against one another—a technology that can improve efficiency, and improved vehicle access and safety that will mean more travel. “It’s a question of which force will dominate,” he said.
That, ultimately, is more about social will than technology innovation. “Automated cars are not going to free us from our own faulty human desire for more and more,” said DeCicco. He said technology can make our lives better, but also bring new and potentially bigger problems. “That’s the story of the automobile over the last 150 years,” he said. “It doesn’t save us from ourselves.”