Imagine trying to drive at night in a secluded area with the headlights turned off. It's not something anyone should be attempting, though that's the condition Ford tested its self-driving Fusion Hybrid, and it did surprisingly well. With the headlights disabled, the Fusion Hybrid successfully navigated desert roads, highlighting the effectiveness of Ford's LiDAR (Light Detection and Ranging) technology.
Don't worry, Ford isn't out in the wild testing its fleet of cars in dangerous conditions where you might run into one. This latest test took place at Ford's Arizona Proving Ground, a non-public area consisting of nearly 1,500 acres. Using LiDAR, the Fusion Hybrid was able to map out the terrain and "steer flawlessly around winding roads" in what Ford described as pitch black conditions.
"Thanks to LiDAR, the test cars aren’t reliant on the sun shining, nor cameras detecting painted white lines on the asphalt," says Jim McBride, Ford technical leader for autonomous vehicles. "In fact, LiDAR allows autonomous cars to drive just as well in the dark as they do in the light of day."
Ford admits that the ideal approach to autonomous vehicles is to use radar, cameras, and LiDAR rather than rely on just one type of sensor. However, these types of tests show that LiDAR might be the most important of the bunch.
LiDAR surveys the landscape using lasers and creates high-resolution 3D maps. These maps contain information about the road and its markings, geography, topography, and things like signs, buildings, and trees. They're also supplemented with RADAR data. The self-driving vehicle then sends out constant pulses to keep track of its position in real-time.
For this particular test, Ford engineers put on night-vision goggles to observe the Fusion Hybrid, both inside and outside the vehicle.
"Inside the car, I could feel it moving, but when I looked out the window, I only saw darkness," describes Wayne Williams, a Ford research scientist and engineer. "As I rode in the back seat, I was following the car’s progression in real time using computer monitoring. Sure enough, it stayed precisely on track along those winding roads."
It's an encouraging test, one that could help autonomous vehicles eventually transition from limited test runs to commercial options.