We’re all anxiously waiting for the day when driverless cars arrive. It’s a bold idea that has been long promised, but generations have gone by with little in the way of results. Suddenly, we feel we are on the cusp.
The bold idea is about to have bold impact.
But, there are key issues remaining in the driverless car debate. One of the most critical is sensing technology, or how the car “sees.” There are two broadly competing technologies, LIDAR and camera sensors. While both will be used in systems in every car, one and only one will win out as the predominant form of location and surrounding sensors. And that difference could mean hundreds of billions of dollars to the winning technology companies.
As Bill Holstein, automotive writer and expert explains in this exclusive Bold Interview, both technologies have strengths and weaknesses. Both will almost certainly have a place on any autonomous vehicle in the future, but one will be the dominant sensor that the rest of the system will be built around. Right now there is a race to be the first and the best technology to help driverless cars navigate through constantly changing busy surroundings.
This allows cars that are using LIDAR technology primarily, to have camera back-ups for those situations where the camera is superior.
LIDAR is an acronym for laser sensing technology. The device sends our beaks of light, and much like sonar, the time it takes for those light beams to return allows the system to build a 3D view of the surrounding area. It has some advantages. First of all, it is already a digital device, so there is no lag time in converting or interpreting signals. Second, it is more reliable in weather conditions like snow or rain when mud can cover a camera lens. One downfall, it can see the streetlight, but it can’t tell if it is lit, it can only sense the location and shape.
Holstein believes that LIDAR will be the primary sensor on cars in the future. He isn’t alone, Ford has made a huge investment with Velodyne to develop the technology.
Camera sensors can see in more detail, and determine if lights are turned on or off. The disadvantage is processing time and weather. A camera image takes time to process to identify what is being seen, and when cars at traveling at typical highway speeds, split seconds are crucial. This gives LIDAR quite an edge. Also, cameras are easily covered with dirt, dust, and snow, which makes them less than optimum in some conditions. Tesla, Mobil-eye and Intel are working primarily with camera sensors.
Both LIDAR and Cameras Have Their Advantages
Because each of the technologies is imperfect, yet powerful, Holstein says that automobiles will almost certainly use both. One of the technologies will be the leader, with a variety of backup systems. This allows cars that are using LIDAR technology primarily, to have camera back-ups for those situations where the camera is superior.
Autonomous cars will have many sensors on any given vehicle. It isn’t difficult to imagine that these will be made up of a combination of technologies and purposes. LIDAR and cameras will certainly work together.
But the leader of the pack will take the grand prize. It’s a race that will have a huge impact on the world.