Startup Adopts 3-Pronged Approach for LiDAR

Article By : Junko Yoshida

Automotive lidar tech supplier Lumotive unveiled three-pronged pivot, by adding smartphone lidar chips to its roadmap...

Lumotive, a Seattle-based startup, is making three-pronged pivot, by adding smartphone lidar chips to its roadmap.

Lumotive announced that late this year it will sample its new lidar products, X20 for automotive and Z20 for industrial applications. Both are based on the light-bending properties of metamaterials that enable the startup’s lidar systems to steer light without relying on the mechanical spinners used by conventional lidars.

Also revealed in Tuesday’s product announcement is Lumotive’s ambitious plan to enter the smartphone lidar sensor market. The company is now adding to its repertoire M20 lidars designed for mobile, with plans to prototype them early next year, with sampling projected for the third quarter.

(Source: Lumotive)

Given the industry’s excitement over Apple’s new iPad Pro lidar sensor, Lumotive’s sudden interest in the mobile segment appears both opportunistic and well justified. During an interview with EE Times, co-founder and CEO Bill Colleran explained that the Lumotive’s solid-state, silicon beam-steering systems are uniquely suited for extreme scalability.

“Our lidar solution is CMOS-based. We can leverage the architecture identical to our lidar solutions in automotive and industrial applications to development of M20 for smartphones,” he said.

Armed with beam-steering lidars based on its liquid crystal metasurface (LCM) chip technology, Lumotive is proceeding on the proposition that its technique can prove superior to spinning assemblies, MEMS-based optics or optical phased arrays. Lumotive’s lidar systems combine a large optical aperture that delivers long range, a 120-degree field-of-view with high angular resolution, and rapid, random-access beam steering.

More specifically, Lumotive pointed out that LCM chips, using tunable sub-wavelength elements based on metamaterials principles, make Lumotive’s lidar systems “more reliable and more compact,” compared to traditional mechanical spinning lidars. Further, the new beam-steering technology can offer “larger aperture for greater range” than MEMS-based lidar systems.

Double whammy
Lumotive’s Colleran expects some shakeout among players later this year in the once over-heated lidar market.

While he appears confident that Lumotive’s differentiated technology can get it through the turbulent times, the basic tenet of Lumotive’s story varies little from other tech startups who joined the AV fray because of what looked like a red hot market.

The initial scenario plotted by the AV industry — anticipating market growth for lidars — indicated that demand would start with a robo-taxi segment where money seemed no object. Then the boom would sweep up ADAS.

In the last eight to ten months, however, the sequence has “flipped,” Colleran acknowledged. Given the industry’s current reckoning that fully automated vehicles will take a lot longer than originally anticipated, the first go-to market for lidars is now in ADAS or industrial applications — not self-driving cars. That reversal demands that lidar technology suppliers make their systems available at a much more aggressive price point, while meeting with different performance.

The switcheroo from AV-to-ADAS to ADAS-to-AV took place first. Then, came the pandemic.

This double whammy of changing market dynamics is forcing a pivot by practically every tech startup. For automotive sensor technology suppliers, the focus has shifted from “let’s dominate the fully autonomous vehicle market” to “let’s find out where else our technology can be applied.”

New products
Sampling in the fourth quarter of 2020, the Lumotive X20 and Z20 build on the company’s X10 prototype of the same architecture, according to the company.

While X10 was billed as a proof-of-concept, both X20 and Z20 are the company’s first “go-to-market” products.

Both lidar systems incorporate modules to transmit, receive and beam-steer. Tier ones or industrial sensor product suppliers can purchase any of the modules to build and improve their own lidar products.

The X20 targets long-range automotive applications with range over 120 meters in bright sunlight and a 120° x 30° field of view. The Z20 will have a shorter range (roughly 50 meters) but an expanded 70° vertical field of view to address industrial and short-range automotive needs.

While the specs of the M20 for smartphones remain sketchy, three building blocks — a laser, LCM beam-steering and a detector – used in the smartphone lidar are based on similar architecture deployed in automotive or industrial systems. For the laser, mobile 3D sensor will use low-power VCSELs (Vertical-Cavity Surface-Emitting Lasers) instead of a high-power diode. For the detector, a small SPAD (Single-Photon Avalanche Diode) arrays replaces the large array used in automotive. The LCM beam steering module must be substantially scaled down to 4mm x 7 to 8 mm, explained Colleran.

(Source: Lumotive)

Colleran believes the Lumotive’s lidar system’s ability to function in bright sunlight will be advantageous when the smartphone incorporated with M20 ventures outdoors. Because Lumotive intends to work with tech suppliers in the lidar ecosystem, Colleran said that its business model remains flexible by supplying its beam steering module, for example, to companies such as STMicroelectronics or Ams who are developing their own SPAD arrays or low-power VCSELs.

The three markets – industrial, automotive and mobile – offer very different pictures of market size and optimum price points. Colleran estimated, by 2025, the mobile lidar market will be as big as $8 billion, with each smartphone lidar system sold at around $8. In comparison, he estimates the size of the industrial market at $2 billion (5 million units) with each lidar at $400. In the same period, he pegged the automotive market at around $4 billion (25 million units) with lidar systems going for $160.

Lumotive is also working on its third-generation X30 lidar X30, to be sampled late in 2021 for the ADAS market. The difference between X20 and X30? X20 is based on FPGAs. X30 will use ASICs.

Leave a comment