Ford Creates Industry First by Conducting Snow Tests of Self-Driving Cars

Ford has created an industry first by conducting snow tests of self-driving cars. This is a major milestone in the company’s drive to became the first automotive manufacturer to take fully autonomous cars to the masses on a global basis.

Many companies, both from the automotive and technology sector have logged thousands of miles in autonomous vehicles. However, these tests have been conducted in dry and mostly sunny conditions. Ford took into consideration the fact that motorists will need to use these vehicles in all kinds of conditions.

Commenting on the tests, Jim McBride, Ford technical leader for autonomous vehicles said that while it might be smooth sailing for a self-driving car in ideal weather conditions, the car’s sensors will be unable to see the road in snowy weather. This makes it crucial for such cars to be tested in winter conditions before they reach the retail scene in view of the fact that almost 70 per cent of US motorists live in areas that experience snow.

Ford will be conducting the winter tests at the Mcity testing facility at the University of Michigan, which covers 32 acres and is a full scale replica of real-world urban conditions.

GPS cannot be used in winter conditions as it offers accuracy only upto several yards and this is not adequate to correctly identify the position of the vehicle. For an autonomous vehicle, it is important to identify its precise location in order to continue on its journey, right down to the driving lane as even variation of a few inches can lead to fatal accidents.

LiDAR, is a lot more effective than GPS as it can help in identifying the location to within a centimeter. LiDAR essentially involves the emission of short pulses of laser light to permit the vehicle to create a real-time, high-definition 3D image of its surroundings.

In dry and sunny weather, LiDAR is the best way to gather data and metadata about the surroundings by sensing objects in the vicinity and using this information to choose the best route. In situations involving snow or high-density traffic, though, sensors like cameras and LIDAR cannot be used to see the road as the snow, debris and traffic will obscure the lens of the sensors.

This challenge drove Ford to partner with the University of Michigan technologists on developing a solution to help autonomous vehicle negotiate snowy roads.

In order to successfully tackle snowy roads, the self-driving cars made by Ford have been equipped with highly detailed high-resolution 3D maps which cover all road signs, landmarks and topography. The maps were created in association with the University of Michigan college of engineering as the maps created by other companies do not work in snowy scenarios. These maps are created when the vehicles negotiate the test environment in good weather with technologies to capture information about identifying features like trees, buildings and traffic signs. Hence, when the vehicle is unable to drive by seeing the ground, it uses these landmarks to identify its position on the map and proceed accordingly.

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