It is impressive to watch a car drive itself. The steering wheel turns without anyone touching it, the car speeds up and slows down depending on traffic, and it can find a spot and park with just a push of a button. It's perfect – partly because it's sunny, dry and warm and conditions are great for a Sunday drive.
However, this often isn't the case in Canada. Even though the technology is advancing quickly, possibly the biggest technical challenge to bringing self-driving cars to market is that sensors can't see through snow.
"It's one thing for a car to drive itself in perfect weather," said Jim McBride, Ford technical leader for autonomous vehicles in a statement. "It's quite another to do so when the car's sensors can't see the road because it's covered in snow."
But progress is being made. Ford announced last week at the Detroit auto show it has figured out a solution and it has begun testing self-driving cars in the snow at Mcity, a 32-acre fake city built to test technology in Michigan. The project is dubbed "snowtonomy" and the tests in wintry conditions are a first for the auto industry.
A lot of semi-autonomous vehicles rely on sensors seeing a 4-inch strip of white paint. Under sunny skies, Ford's autonomous vehicles rely on LiDAR sensors, which can pinpoint the vehicle's location in a lane down to the centimetre. GPS can only locate a car within a few metres. The sensors are the spinning objects on top of the car. They emit short pulses of laser light, which create a real-time, high-definition 3D map of the surrounding area. The sensors have to be able to see the lane markings and can't be covered by snow or grime in order to work.
Ford and University of Michigan engineers are working on a solution that involves 3D maps that include information about what's above the road such as landmarks, signs, buildings and topography.
"Maps developed by other companies don't always work in snow-covered landscapes," said Ryan Eustice, associate professor at the University of Michigan college of engineering, in a statement. Mcity is part of the university. "The maps we created with Ford contain useful information about the 3D environment around the car, allowing the vehicle to localize even with a blanket of snow covering the ground."
The maps are created during favourable conditions meaning a self-driving car can't navigate a road for the first time if it is covered in snow. However, after driving a route once, the self-driving car can then navigate the route when covered in snow by using the above ground features.
Ford did the testing in a Fusion Hybrid and said the electronic stability control and traction control worked in unison with the autonomous driving software.
A few days ago, Google released its monthly self-driving car report for December, which describes testing in recent rains.
"Our cars can determine the severity of the rain, and just like human drivers they drive more cautiously in wet conditions when roads are slippery and visibility is poor," reads the report. "To explore even more challenging environments, we're beginning to collect data in all sorts of rainy and snowy conditions as we work toward the goal of a self-driving car what will be able to drive come rain, hail, snow or shine."
Like us on Facebook
Add us to your circles
Sign up for our weekly newsletter.