UC Berkeley researchers enhance MEMS-based LiDAR

“Normal cameras involve a 2D sensor sensing an image independent of distance and compressed on a plane,” Wu said. “For many applications, from self-driving cars to robotics to drone navigation, it’s important to have that 3D landscape to avoid obstacles and plan their path.”

The study's co-first authors, Xiaosheng Zhang and Kyungmok Kwon, stressed the need of shrinking the size of LiDAR while retaining its performance.

According to Zhang and Kwon, one of the most difficult aspects of this undertaking is creating an integrated beam scanner that "steers" the output laser light in multiple directions to scan throughout the area. The spinning item on the roof of self-driving automobiles is the LiDAR beam scanner.

Instead of more usual thermo-optic switches, the group's approach employs small microelectromechanical system, or MEMS, switches. According to Zhang and Kwon's email, MEMS switches physically move the waveguide, a tube that channels electromagnetic waves, up and down to route light to each pixel.

This reduces the pixel size from several hundred micrometers to 50 micrometers, according to Wu. The researchers were thus able to increase the number of pixels on a one-centimeter-squared semiconductor from 512 to 16,384 while using much less power.

“Using similar silicon technology, we are on the path of scaling the technology and can reach even higher resolutions in the future.”, Wu said.

According to Wu, LiDAR is only one of several systems used to identify objects, particularly in self-driving vehicles, thus the sensors must operate well together.

For example, if a standalone sensor cannot tell the difference between a vehicle and the blue sky, LiDAR may give important information when properly integrated.

“Making critical components more accessible and cheaper will enable it to become ubiquitous,” Wu said. “Digital cameras used to be very expensive … but very high-performance cameras are now very cheap to the point that people will not hesitate to add a camera anywhere they need it.”

Zhang and Kwon saw self-driving vehicles being tested in San Francisco on a regular basis, several of which used LiDAR. They want to improve self-driving vehicles and LiDAR's many other uses by enhancing this technology.

The next challenge for Wu's team is getting this technology from the lab to the market, which entails figuring out how to make the gadget replicable in big numbers.

A large-scale microelectromechanical-systems-based silicon photonics LiDAR - Nature

A large focal plane switch array is constructed to steer the laser beam of a LiDAR system, leading to 3D imaging with 16,384 pixels, improving the resolution and coverage of solid-state LiDARs.

NatureXiaosheng Zhang