A driverless car’s delightful—unless the weather outside is frightful
▶ “Snow smoke” disables sensors robot vehicles need to navigate ▶ “When it starts building up, you just lose functionality”
In Jokkmokk, a hamlet just north of the Arctic Circle in Sweden, Volvo Cars’ self- driving XC90 sport-utility vehicle met its match: frozen flakes that caked on radar sensors essential to reading the road. Suddenly the SUV was blind. “It’s really difficult, especially when you have the snow smoke from the car in front,” says Marcus Rothoff, director of Volvo’s autonomous-driving program. “A bit of ice you can manage. But when it starts building up, you just lose functionality.” So Volvo engineers started moving the sensors. And next year, when Swedish drivers take their hands off the wheel of leased XC90S in the world’s first public test of fully autonomous technology, the radar will be nestled behind the windshield, which wipers keep clear of ice and snow.
As automakers race to get robot cars on the road, they’re encountering an obstacle very familiar to humans: Old Man Winter. Simple snow can render the most advanced computing power useless. That’s why companies including Volvo Cars, owned by China’s Zhejiang Geely Holding Group, and Ford are revving up efforts to prevent snow blindness.
With about 70 percent of Americans living in the snow belt, learning to navigate in rough weather is crucial for driverless cars to gain mass appeal and realize their potential to reduce traffic congestion and road deaths. “If your vision is obscured as a human in strong flurries, then vision sensors are going to encounter the exact same obstacles,” says Jeremy Carlson, an autonomy specialist at IHS Automotive.
Driverless cars “see” the world around them using data from cameras, radar, and lidar, which bounces laser light off objects to assess their shape and location. High-speed processors crunch the data to provide 360-degree detection of lanes, traffic, pedestrians, signs, stoplights, and anything else in the vehicle’s path. But snow can shroud cameras and cover the lane lines they must detect to keep a driverless car on course. Lidar also is limited because the light pulses it emits reflect off flakes, potentially confusing even a modest curtain of falling snow with an object to avoid, causing the car to hit the brakes.
Radar, which senses objects by emitting electromagnetic waves and has been used since 1999 in adaptive cruise control to maintain a set distance from other vehicles, is better. “If everything else fails, I can follow the preceding traffic,” says Kay Stepper, vice president and head of the automated driving unit at German parts supplier Robert Bosch. “The radar is the key
element of that because of its ability to work robustly in inclement weather.”
Ford, which declined interview requests, says it might have found a solution to snow-blanketed lane lines, according to a press release. The automaker is scanning roads in advance with lidar to create high-definition 3D maps that are more accurate than images from global positioning satellites, which can be 10 meters off.
Ryan Eustice, an associate professor of engineering at the University of Michigan who’s worked with Ford on the problem since 2012, says they’ve also found a way to filter the “noise” created by falling snowflakes. The filtered data combined with information from the 3D maps allow the car to pinpoint its location to within “tens of centimeters,” he says. “That’s high enough accuracy that we know exactly what lane we’re in.”
Still, lane lines can become meaningless in a snowstorm, as human drivers blaze their own trails in the ruts created by vehicles in front of them. The solution may be artificial intelligence, says Danny Shapiro, senior director of automotive at processor maker Nvidia, which says its latest computer brain can perform as many as 24 trillion “deep learning operations” per second. Deep learning trains a robot car, based on millions of miles of driving experience loaded into its software and continually updated. So in a snowstorm, the car will know it should follow the ruts rather than stay within the lane lines. “It’s very similar to how a human learns, by experience,” Shapiro says.
Also like a human, a driverless car can get disoriented in a whiteout. “There’s been a lot of hype in the media and in the public mind’s eye” about the technology for selfdriving cars “being nearly solved,” Eustice says. “But a car that’s able to do nationwide, all-weather driving, under all conditions, that’s still the holy grail.” The bottom line About 70 percent of Americans live in the snow belt. So carmakers are trying to develop driverless vehicles that can “see” in snow.