Lumotive, a Seattle-based, venture-backed startup, is unveiling Wednesday a lidar technology based on metamaterials, a relatively new and exotic technological approach that few if any of its competitors have adopted.
Among the many sensors destined to go inside advanced driver assistance systems (ADAS) and autonomous vehicles (AV), lidars continue to be a hotbed of new technology. The flip side of this vibrant activity, however, is the hard reality that the lidar market remains embryonic and its technologies fragmented. With more than $800 million pouring into lidar tech startups over the last few years, Alexis Debray, technology and market analyst at Yole Développement (Lyon, France), estimated, “There are 60 to 70 lidar companies that have popped up worldwide.”
Lumotive is jumping in to this crowded field with the expectation that by using metamaterials, it can surpass its more established rivals. Lumotive has its own version of the technology, which it describes as liquid crystal metasurfaces (LCM).
In an interview with EE Times, Lumotive co-founder and CEO Bill Colleran called metamaterials a “pivotal technology” that opened the door for Lumotive to develop a lidar that can steer beams with no moving parts. Based on the light-bending properties of metamaterials, LCM can steer light without relying on the mechanical spinners in conventional lidars, including those used by Waymo.
Of course, Lumotive isn’t the first to highlight a lidar that doesn’t rely on mechanical scanning. A host of rivals utilizing MEMS mirrors or optical phased arrays have already claimed their lidars use fewer components or “no mechanical parts.”
Colleran, however, stressed, “Ours are different,” because its LCM semiconductor chips come with a larger optical aperture (25 x 25 mm). This allows a lidar with a much longer range. Combining a 120-degree field of view with fast random-access beam steering will enable Lumotive’s system to offer high performance, the startup promised.
The image shows a laser shining onto Lumotive's Liquid Crystal Metasurface (LCM) chip. By programming electrical signals onto the chip, the reflected light is directed into any direction over a 120-degree field of view.” (Source: Lumotive)
Yole analyst Debray, who was briefed by Lumotive, called the technology “original,” and “interesting.” However, since there is no system prototype yet available, he described the product “still preliminary and in early stage.”
Asked about a launch schedule, the Lumotive CEO told us that samples will be available in the third quarter this year. Will it be automotive-grade qualified? Colleran said that is more than two years away. However, he appears cool with the late schedule because there are opportunities for his company’s lidars beyond automotive — including industrial, robotics and drones. Lumotive also expects its lidars to be used as after-market options (which are not required to be automotive qualified). Lumotive appears to be playing a long game, expecting to intersect with a volume automotive market at a lower cost, which Colleran sees still several years off.
In Debray’s view, the lidar market – where investment activities have overheated – will evolve in “phases.” The first wave is already happening in a market crowded with traditional mechanical lidars. Debray sees the second wave starting this year with lidars based on MEMS. These will be smaller and cheaper than previous generation lidars. Optical phase-array lidars, flash lidars and something entirely different like Lumotive’s LCM-based lidars will be the third wave, he predicted.
The story of Lumotive can’t be told without referring to metamaterials.
Metamaterials are defined as “artificially structured materials.” The advantage of metamaterials is the opportunity to exert unprecedented control over their properties or a related device. A metamaterial is not a specific thing, but rather “a design approach” that can unleash new ways to control the flow of light and other wave excitations.
A good example of the commercial use of metamaterials is Kymeta’s flat-panel antenna technology, explained David Smith, professor of physics at Duke University and a pioneer in metamaterials research. Kymeta has commercialized a new metamaterials-based satellite antenna.
Kymeta leverages the “particular architecture” of metamaterials, he noted. Its antenna can move electronically without phase shifters, related amplifiers and other components at each point of its surface. Too many phase shifters and amplifiers “would cost money, burn power and require cooling devices,” said Smith.
By applying metamaterials design principles, Kymeta was able to dramatically increase the density of antenna elements in its flat-panel antenna, compared to a phased array. Phase and amplitude can be controlled simply by activating or deactivating antenna elements.
The metamaterial architecture applied to Kymeta’s antenna has been adapted to Lumotive’s lidar. The difference is that while metasurface concepts have been applied to microwave frequencies in the past, this is the first use of dynamic metamaterials developed for optics, Smith explained.
Lumotive’s use of artificially structured surfaces to control electromagnetic waves is unique. “But the use of metamaterials for optics involved a couple of issues,” said Smith. First, the competition in optics is “a lot tougher,” he noted, since optical phased array lidars are already in development. “But the problems with phased array-based lidars include their density and heat,” he explained. “That left us an opportunity for metasurfaces.”
Commercialization of metamaterials
Discovering metamaterials is one thing. Commercializing the technology is a whole different story.
Smith, best known for his theoretical and experimental work on electromagnetic metamaterials, has been a pioneer in revealing the potential of metamaterials. He has worked closely with Invention Science Fund (ISF), an investment fund of Intellectual Ventures. A partnership founded in 2000 by Nathan Myhrvold and Edward Jung of Microsoft, Intellectual Ventures is one of the top five owners of U.S. patents (and is also a well-known patent troll).
While others are also working to commercialize metamaterials, Smith said that because Intellectual Ventures possesses the world’s leading portfolio of metamaterials IP and ISF has its own startup arm called ISF Incubator, it is “most likely to be able to capitalize on the technology. They are actually doing something with it.”
Lumotive, which began its work inside the ISF incubator, has an “exclusive license” of the particular metamaterial architecture for specific application in lidars. Similar to Kymeta, Lumotive has spun out of Intellectual Ventures.
As Lumotive explains, the LCM chip steers the laser light using tunable sub-wavelength elements based on metamaterials principles. The startup is confident that it can make its LCM chip-based lidar “reliable, cost effective and mass production-ready” because it uses a semiconductor manufacturing process.
Keymeta, for example, is using Sharp as its foundry. What about Lumotive?
Colleran told us, “We’ve used a couple of well-known foundries over the past year and a half.” While indicating plans to focus on one foundry as a lead, he declined to name a foundry partner at this point.
Today, mainstream lidars are based on mechanical spinning assemblies. Waymo, Alphabet Inc.’s self-driving car unit, announced this month that it will start selling its home-grown lidar, Laser Bear Honeycomb, to companies who agree not to compete with Waymo’s self-driving taxi plans.
At the time of its announcement, Waymo said that it is making these sensors available to vendors outside of self-driving — such as robotics, security and agricultural technology. Waymo’s Laser Bear Honeycombs are much smaller than the current-generation lidars spinning on the roof of a self-driving car, but their basic technologies remain mechanical.
Ford Motor Co.’s Argo AI unit also has its own lidars – developed by New Jersey-based Princeton Lightwave, which Argo AI acquired in 2017. Princeton Lightwave is known as a developer of a flash lidar called “Geiger-mode LiDAR technology” that detects and processes photons digitally in real time.
Leading developers of robotaxi technologies already have their in-house lidar technologies. So how does a startup like Lumotive break into the market?
Colleran told us, “First, it’s good news. Because it implies the industry’s unmet needs – prompting these leading companies to develop lidars on their own.” Of course, the bad news is that these robotaxi pioneers are no longer Lumotive’s potential customers, he acknowledged.
Yole’s Debray said that for many lidar startups to survive, their priority is to find partnerships with Tier Ones. “They need to be a part of the supply chain. They must establish close collaborations with them.”
Lumotive, which declined to comment on the size of its seed money, said it will be seeking its first round of financing later this year.