Most automated vehicles detect objects around them using a combination of cameras, radar, and lidar. They each have different types of sensors and each of them has weak points. Highly sophisticated software is needed to distribute the workload and consolidate the data obtained through each type of sensor into one complete picture.
A startup called AEye has developed a hybrid sensor that it claims can streamline the whole process by combining a solid-state lidar unit, low-light camera, and chips which can run AI algorithms. AEye was set up Luis Dussan, who has worked at renowned companies like Northrop Grumman, Lockheed Martin, and NASA’s Jet Propulsion Laboratory. He started out by focusing on AI for self-driving cars, and went on to realize that the sensors that are currently available did not have the capability to provide the kind of data he needed.
Using AEye’s hybrid “iDAR” sensor, which is a solid-state lidar unit can help keep costs down as it is cheaper than conventional lidar units. However, it is not as effective as conventional lidar as it cannot t aim light beams across an area that is as broad. The company is hoping to balance this by redirecting the beams. Its lidar unit can scan a wide area at low resolution, or a specific spot at high resolution and similar to the way a human eye functions, it can shift its focus depending on what is needed.
The key advantage of using this hybrid sensor is the way in which the lidar and camera complement each other. The camera can add color to the lidar unit’s images, and this is crucial for identifying features like brake lights. It can also tell the lidar where to focus on, with the help of image-recognition software to prioritize objects. AEye says the result is better with this kind of selective approach, rather than scanning everywhere all of the time.
The AEye iDAR sensor does have some disadvantages. The sensor’s field of view is only 70 degrees, and hence multiple sensors need to be placed on a vehicle to get full coverage. Price is also likely to be quite high as the emphasis is more on performance than on cost.
