Elon Musk doesn’t care for Lidar — the light-based echo location technology that is similar in concept to radar — even though Volvo says it will incorporate Lidar units from Luminar into some its cars by 2022. The big difference is, radar uses radio waves, which are of much lower frequency than the light waves used by Lidar. Those lower frequency waves are quite good at penetrating fog, smog, haze, rain, snow, and drizzle. Light waves do not do such a good job when such atmospheric obstructions are present.
What Lidar does better than radar is create more accurate representations of objects in its path. That’s good. But current units are expensive, costing up to $75,000 each. That’s bad. What makes it worse is that Lidar units tend to have a relatively narrow field of view, which means several units are needed to guide a self-driving car along its intended path safely.
Imagine there was such a thing as a wide angle Lidar unit that cost only a few hundred dollars. That would really move the quest for self-driving cars forward, wouldn’t it? (And make a lot of money for the inventor, presumably.) Two groups of researchers say they are closing in on that goal. Here’s what’s new in Lidar these days.
We are going to start today’s discussion with Toshiba, not because its Lidar research is any better than anyone else’s, but it is a recognized electronics manufacturer with decades of experience at manufacturing actual stuff. Its announcement is titled “Advanced Light-Receiving Technology from Toshiba Enables Solid-State LiDAR Free from Reliance on Mechanical Components.” Toshiba says its new technology is compact, low cost, and simple to install, making it suitable for use in a variety of automotive applications.
“By removing the need for bulky mechanical components, the technology realizes cost and space savings and enhances operational reliability. At its heart is Toshiba’s proprietary compact, high-efficiency silicon photo-multiplier (SiPM). In general, SiPM are suitable for long-range measurement as they are highly light sensitive.
“However, the light-receiving cells composed on SiPM require recovery time after being triggered, and in strong ambient light condition they also need a large number of cells, since they must have reserve cells to react to reflected laser light. Toshiba’s SiPM applies a transistor circuit that reboots the cells to reduce the recovery time. The cells function more efficiently and fewer are needed, securing a smaller SiPM, as shown in Figure 1. This realizes a higher-resolution SiPM array while maintaining high sensitivity.”
See graphics below.
“Field trials with a LiDAR prototype using commercially available lenses, from wide-angle to telephoto lenses, have demonstrated the system’s effectiveness over a maximum distance of 200 meters. This effectively quadruples the capabilities of currently available solid-state LiDAR systems while maintaining high resolution.”
The use of commercially available lenses “eliminates complex customization and allows it to be applied in a wide variety of self-driving vehicles. The technology’s inherent compactness allows placement in multiple locations on vehicles that had previously presented a major challenge; and it will also broaden LiDAR applications outside of the automotive sector.” Toshiba plans to begin manufacturing its new compact Lidar units in 2022.
Please watch the video below for more on this topic.
University of Colorado Boulder
Researchers at the University of Colorado Boulder also say they have made a breakthrough in Lidar technology. In a paper published in the journal Optica with the catchy title “Serpentine optical phased arrays for scalable integrated photonic lidar beam steering,” the researchers say they have been working on a new way of steering laser beams called wavelength steering — where each wavelength, or “color,” of the laser is pointed to a unique angle.
According to CU Boulder, they have not only developed a way to do a version of this along two dimensions simultaneously, instead of only one, they’ve done it with color, using a “rainbow” pattern to take 3-D images. Since the beams are easily controlled by simply changing colors, multiple phased arrays can be controlled simultaneously to create a bigger aperture and a higher resolution image. “We’ve figured out how to put this two-dimensional rainbow into a little teeny chip,” said Kelvin Wagner, co-author of the new study and professor of electrical and computer engineering.
“Electrical communication is at its absolute limit. Optics has to come into play and that’s why all these big players are committed to making the silicon photonics technology industrially viable,” said Miloš Popovic, co-author and associate professor of engineering at Boston University. The simpler and smaller that these silicon chips can be made — while retaining high resolution and accuracy in their imaging — the more technologies they can be applied to, including self-driving cars and smartphones. There is a rumor the iPhone 12 may include Lidar for enhanced facial recognition.
“We’re proposing a scalable approach to lidar using chip technology. And this is the first step, the first building block of that approach,” says Nathan Dostart, lead author of the research study. He will continue his research at NASA’s Langley Research Center in Virginia. “There’s still a long way to go.”
Several companies. including Bosch, say they have advanced Lidar technology for cars ready to go. And researchers at Stanford say they are making good progress on Lidar-on-a-chip technology. Still, stuff that works in the lab can take many years to achieve commercial success. Right now, based on this latest announcement, it appears Toshiba may be in the lead when it comes to making Lidar technology affordable and scalable.
Autonomous vehicles are expected to be a $500 billion per year business within 5 years, but the technology to make fully self-driving vehicles that are not constrained by the need to follow predetermined routes is not yet available at an affordable price. Perhaps one of these research announcements will solve that issue.