As automakers continue to crack the code for a perfect self-driving vehicle, several methods are being devised to reach the endpoint. A majority of automakers including old-timers such as Audi have preferred Lidar over other solutions such as end-to-end machine-learning approach and just cameras and radar. All of it boils down to one simple question – which is the safest of them all?
What is Lidar?
In simple words, Lidar is a method in which distance to a target can be measured by illuminating the target with pulsed laser light and measuring the reflected pules with a sensor. Differences in laser return times and wavelengths can then be used to make digital 3-D representations of the target.
Apart from autonomous vehicles, Lidar is commonly used to make high-resolution maps, with applications in archaeology, geography, geology, geomorphology, seismology, forestry, atmospheric physics, laser guidance, airborne laser swath mapping (ALSM), and laser altimetry.
Aeva will be responsible to equip the test fleet of Audi E-Tron vehicles with its 4D Lidar.
Both technologies are efficient in their own approach. However, it appears that German auto-giant Audi is firmly on Team Lidar. And for good reasons indeed.
On Wednesday, Aeva announced a strategic partnership with Autonomous Intelligent Driving, a research unit that is wholly owned by Audi. The department is responsible for developing AV tech for the manufacturer and plans to equip the test fleet of Audi E-Tron vehicles with its 4D Lidar.
Over the past few months, Aeva has been a discussion in the automotive industry for its technology. The company has developed a Lidar system that can detect the velocity of an object in a single frame at up to 300 meters. A typical solid-state Lidar uses an array of lasers to build a point cloud of objects in its path, measuring the speed and direction of an object by calculating movement between frames. However, Aeva is able to skip the calculation process using only a single beam and coded signals that detect frequency changes to calculate velocity. This feature cuts time and frees up those milliseconds so they can be better spend making decisions earlier.
Aeva's Lidar system is able to skip the calculation process using only a single beam and coded signals that detect frequency changes to calculate velocity.
In addition to this, Aeva’s 4D design makes it immune to interference. While this might not be an issue in the current scenario, a future with thousands of vehicles, all equipped with Lidar passing each other on the road could cause false signals.
Many automakers around the world have made up their minds that Lidar will be a part of their autonomous driving tech. “We believe that for a safe stack you must have all the outputs to make the right decision,” says Aeva co-founder Soroush Salehian. “If you’re using just a camera sensor, there are so many issues that may blind the system. You need to be able to compliment the different sensor technology weakness–get both outputs and its up to the carmaker to decide what to do with that information. What we provide is the ability to see what they couldn’t see before and measure things directly.”