It’s a steadfast rule of the universe: Energy can neither be created nor destroyed, merely transferred. And it seems the same idea applies to efficiency, especially when it comes to self-driving cars.
While the robo-vehicles of tomorrow will no doubt make far better drivers than their organic counterparts, generating all the mechanical brainpower needed to do so will take a little too much tangible fuel.
As the autos are asked to take on more responsibility, they require more energy—sucking up as much as 100 laptops’ worth of juice to complete increasingly intricate computations. But national requirements—and, indeed, global trends in the automotive world—demand that future cars run on cleaner fuel than ever before, and start seriously stepping up their miles-per-gallon ratios.
With deadlines for commercial release of the vehicles quickly approaching, automakers have a limited time to pull off the engineering acrobatics needed to make the cars drive themselves on reduced power. And some experts are worried the elements necessary to produce vehicles that stand a chance at making the efficiency cut could soon be in short supply.
Still, industry demand for autonomous autos has never been higher—but will we run out of ways to make the cars run?
Hungry for More
In the age of electronics, the power outlet has become an increasingly important design feature. Places to plug in have popped up everywhere from car dashboards to airplane seats, and the appearance of an outlet in any public arena is always sure to draw a crowd of low-batteried travelers waiting their turn for a charge.
In 2014 alone, powering all of America’s Internet servers required 70 billion kilowatt hours—or the equivalent of running 6.4 million households for an entire year, according to a study by the Department of Energy’s Lawrence Berkeley National Laboratory. (That figure is also projected to rise steadily and quickly, up to as much as 140 billion kilowatt hours—or the additional output of 17 power plants—by 2020, the National Resource Defense Council found.)
While self-driving cars don’t need nearly as much fuel to get going—each vehicle uses between 2 to 4 kilowatt hours of juice—the more automated they become, the more energy they require, demanding more power from their already maxed-out batteries.
All told, it will be as much as 10% more difficult for an autonomous auto to meet future emissions standards, compared to their human-driven counterparts, said Chris Thomas, chief technology officer at propulsion system supplier BorgWarner Inc. While that figure is based off of prototypes, and the technology is likely to become more streamlined as time goes on, Thomas told Automotive News that even if software manufacturers were able to fulfill their ambitious promise of cutting power consumption by 90%, engineers would still have to look for energy gains elsewhere to keep the cars at the legal consumption limit.
Currently, those standards require an average of 54.5 miles per gallon by 2025, with the goal for subcompact cars set at 58.4 mpg and a 48.4 mpg target for small utility vehicles. (The figures have since been reopened for examination by the current presidential administration.)
But a fully autonomous Honda Fit, which falls into the subcompact category, would get no more than 54.6 miles per gallon in a best-case scenario by 2025, Thomas explained, while a self-driving small SUV would optimally see 45.8 miles per gallon by that time.
It may not sound like much of a difference, but in an industry where even the addition of an LED light is fretted over due to its potential impact on efficiency, the gap is “not trivial,” Thomas said.
Gaining all of the knowledge needed to drive hasn’t just made autonomous vehicles hungrier; it’s also made them heavier.
All the equipment the vehicles need to function—including the full suite of cameras, radar sensors, and other technological add-ons—contribute to an even greater total weight requiring still more power to haul around.
Carmakers have been trying to counter this trend even before the rise of autonomous autos, seeking out lighter materials such as aluminum, high-strength steel, carbon fiber, compressed wood, and even soy to build vehicles that could more easily meet emissions standards. And with good reason: For every 10 pounds of weight that can be dropped from a vehicle, 10 to 15 fewer pounds of carbon dioxide will enter the air, according to a report by the Wall Street Journal.
Still, automotive companies must take care not to overuse the lightweight materials. The rise of electric vehicles has already caused copper shortages around the world. Cobalt and lithium deposits are also reportedly in danger of drying up.
Without the proper elements to build stronger batteries, fueling such vehicles could prove extraordinarily difficult—an idea that flies in the face of both the self-prophetic cries of some major auto industry players like Tesla and Lyft, which have both promised entirely electrical futures, and the federal government’s increasingly demanding emission standards.
Weighing Out the Options
While the efficiency problem of self-driving cars is still far from solved, manufacturers have wasted no time putting out prototypes—and many of them are attached to a hybrid engine.
Google affiliate Waymo recently unleashed a fleet of 600 self-driving hybrid minivans in Phoenix, while Jim Farley, president of Ford’s global markets, told investors earlier this month the company believes hybrids are “the right tech to start with” when it comes to autonomous vehicles.
Whether the gas-supported engines can continue to help vehicles pass emissions inspections—and whether they’ll ultimately slow the adoption of fully-electric cars—remains to be seen.
Still, ambitious automakers could take a page from Tesla’s Elon Musk, who’s preparing to unveil an all-electric semi-truck soon. While the vehicle can reportedly only make it 200 miles before needing a recharge, that type of battery power could help haul the entire auto industry in the right direction.