There are scores of lidar startups. Why read about another one?
It wouldn’t be enough to have the backing of a big car company, though Aeva has that. In April the company announced a partnership with Audi, and today it’s doing so with Porsche—the two upscale realms of the Volkswagen auto empire.
Nor is it enough to claim that truly driverless cars are just around the bend. We’ve seen that promise made and broken many times.
What makes this two-year-old company worth a look-see is its technology, which is unusual both for its miniaturization and for the way in which it modulates its laser beam.
Aeva shrank the lidar system a thousandfold by putting all the basic functions onto a single photonics chip. Only the steering of the beams is relegated to a second chip, which uses an array of microelectromechanical mirrors.
“All lasers, detectors, circuitry, are integrated on a chip at the wafer level, and it’s in production for volumes already targeting use in major OEMs,” says Mina Rezk, a cofounder of the company. “It will be in cars in the 2022 to 23 timeline.”
Aeva says mass production should push the unit selling price below US $500. Because the lidar sweeps 120 degrees horizontally and 30 degrees vertically, one or two units should suffice for an ADAS-capable car. A true driverless car would need three or four units, which could push the total cost per car above $1,000. That’s still way below the tens of thousands you’d pay for the best 360-degree lidar offered by Velodyne, the founder of the automotive lidar industry.
Cost has been the main obstacle to wide acceptance of lidar in cars. But VW is emphasizing performance.
“In my view, this lidar is unique because it has all the requirements,” Alex Hitzinger, senior vice president of autonomous driving at VW Group, tells IEEE Spectrum. “It has high resolution, long range, additionally it gives velocity. It’s scaleable because it’s all on a photonics chip–a very small package with low power consumption.”
The mode of modulation differs from that of most other lidars, which emit short pulses and measure the time it takes the reflection to return to infer distance to an object. They must make at least two successive measurements to calculate an object’s speed.
Aeva instead relies on the principle of FM radio: It modulates the frequency of the beams. By scrutinizing the reflections, the system can infer both an object’s position and its instantaneous speed, as revealed by the change in wavelength due to movement. That’s the familiar Doppler shift you hear in the rising and falling pitch a siren makes as it comes and goes.
A few other lidar startups also rely on a version of frequency modulation. But Aeva’s system uses a continuous beam, rather than a pulsed beam.
Soroush Salehian, the other cofounder of the company, says the lidar can resolve objects at distances of up to 300 meters—the outer limit for today’s best lidars. He says the lidar can still detect objects that reflect as little as 10 percent of incident light.
“That’s good for seeing lane markings, or for reading things that are inside the traffic sign—lane merge, things like that,” Salehian says.
VW’s ID Buzz AV Photo-illustration: VW
Aeva’s lidar has been on test vehicles for some time, including an experimental version of Audi’s E-Tron. The first commercial application may come in lightly automated, or Level 2, vehicles, but right now, all VW will say is that it plans to put Aeva’s lidar on the ID Buzz AV. That’s a more completely automated, or Level 4, minibus; it’s scheduled to launch in 2022/2023.
Philip E. Ross is a senior editor at IEEE Spectrum. His interests include transportation, energy storage, AI, and the economic aspects of technology. He has a master's degree in international affairs from Columbia University and another, in journalism, from the University of Michigan.