Can a self-driving car really survive a Canadian winter?

Despite their delusions of superior skill behind the wheel, many Canadians aren’t very good at driving in the snow.

Sure, we like to be condescending to “those idiots” who drive into the ditch or play crash-up derby on the highway in bad weather, but most of us have never been trained to drive in the snow, and it shows.

If testing by auto makers is any indication, the task may soon be taken out of our hands — and that might be a good thing. After all, it’s the driver, not the car, who panics and plows into the snowbank at the side of the road, or slams on the brakes at the first sign of a skid on ice.

Ford has been testing autonomous vehicles at its facility in Michigan and has been working to overcome a significant hurdle, which is making the self-driving car work safely and efficiently in poor weather conditions.

To navigate snowy roads, Ford autonomous vehicles are equipped with high-resolution 3D maps, complete with information about the road and what's above it, including road markings, signs, geography, landmarks and topography. (Photo: Ford)

Can the car handle treacherous winter driving conditions?

“It’s one thing for a car to drive itself in perfect weather,” Ford’s technical leader for autonomous vehicles Jim McBride said in a release. “It’s quite another to do so when the car’s sensors can’t see the road because it’s covered in snow.”

Ford says its winter driving testing of autonomous cars in Michigan is the first of its kind. The Detroit auto maker uses “high-resolution 3D mapping and LiDAR for localization to facilitate driving when road markings are not visible.”

Light Detection and Ranging (LiDAR) is more accurate than GPS, and can place a car within a centimetre in its lane using short pulses of laser light. GPS is only accurate within a few metres.

But does it work when the road is covered with snow?

Not yet. While LiDAR works in clear weather, if it can’t see the road due to snow or high-density traffic, it won’t work. It will also fail if the sensors are covered with snow.

A solution is in the works, however. Ford says its autonomous cars are equipped with high-resolution 3D maps. These maps go beyond the routes found on regular maps, and include road markings, signs, geography, landmarks and topography.

The maps are created as the vehicle drives the route in good weather, allowing the car to situate itself based on surrounding landmarks if the road isn’t visible. Ford has been working on this technology for a decade now.

(Photo: Ford)

The sensors that create the maps are called Solid-State Hybrid Ultra Puck Auto, named for their similarity to a hockey puck’s size. They are attached to the side mirrors and have a range of 200 metres. They will help the development of the virtual driver software, which is the decision-making “brain” in the self-driving cars.

Auto makers already use a variety of sensors and cameras to operate existing features on cars such as backup assist, adaptive cruise control and lane-departure warning.

What about the driving?

Drivers can make mistakes on slippery roads, and so having an unemotional computer making driving decisions could be better.

“The vehicle’s normal safety systems, like electronic stability control and traction control, which often are used on slippery winter roads, work in unison with the autonomous driving software,” said McBride.

Algorithms are used to do a lot of “what-if” planning, so the autonomous car can be trained to perform evasive manoeuvres as well as an expert, highly trained driver, says Greg Stevens, Ford’s Global Manager of Driver Assistance Research.

Your car could be more sensible than you

The auto maker has 30 self-driving vehicles in testing facilities in California, Arizona and Michigan.

“We eventually want our autonomous vehicles to detect deteriorating conditions, decide whether it’s safe to keep driving, and if so, for how long.”