Lumotive unveils LiDAR with LCM beam-steering technology for autonomous vehicles
21 March 2019
Lumotive, a Bill Gates-funded startup developing LiDAR systems for autonomous vehicles, introduced a beam-steering technology which will significantly improve the performance, reliability and cost of LiDAR systems for the emerging self-driving car industry. Initially targeting the robo-taxi market, Lumotive’s patented system uses Liquid Crystal Metasurfaces (LCM) and silicon fabrication to achieve unmatched levels of manufacturing efficiency while simultaneously delivering excellent range, resolution and frame rate.
Lumotive’s new LiDAR system features beam steering technology leveraging Liquid Crystal Metasurfaces and semiconductor manufacturing.
In the Lumotive LiDAR system, a laser beam is incident on the reflective semiconductor chip and is reflected in a programmable direction, depending on the electronic configuration of the metamaterial elements on the surface of the chip. The beam can be pointed in any direction, in any sequence.
Our LiDAR sensors benefit tremendously from the unique attributes of beam-steering LCMs which simultaneously offer large optical aperture, wide field-of-view and fast scanning while having no moving parts. LCMs deliver the combination of performance and commercial viability that will finally eliminate barriers to adoption of LiDAR for both ADAS and autonomous vehicles.
—Lumotive co-founder and CEO, Dr. William Colleran
LiDAR has emerged as a key 3D sensing technology by enabling autonomous driving systems as well as Advanced Driver Assistance Systems (ADAS). LiDAR systems determine range by emitting laser light pulses and measuring the round-trip flight time for those pulses to travel to and reflect back from objects. A LiDAR system creates a 3D perceptual map, or “point cloud,” of its surroundings by scanning, or “beam steering,” laser pulses across its two-dimensional field-of-view, with the third dimension derived from the distance measured to an object at a given horizontal and vertical position.
LiDAR has proven critical for autonomous vehicles because the technology can accurately locate objects to within a few inches at ranges of hundreds of yards. LiDAR, which stands for Light Detection and Ranging, generally exhibits shorter range but superior measurement resolution compared to its older cousin, radar, which stands for Radio Detection and Ranging.
Currently, most LiDAR systems rely on mechanical scanning, which suffers from poor reliability, cost and form factor and, most importantly, limits the performance of existing systems. In contrast, Lumotive’s beam-steering technology uses LCMs—semiconductor chips that steer laser pulses based on the light-bending principles of metamaterials, a first for LiDAR. Lumotive’s LCMs have large apertures to improve LiDAR perception while benefiting from the economics of semiconductor manufacturing to enable low cost systems. Initial production units will be available to select customers for beta testing in the third quarter of 2019.
Beam steering is the missing link required for high performance and commercially viable LiDAR systems. Traditionally, LiDAR relied upon unreliable and bulky spinning assemblies, while newer breeds of LiDAR sensors utilize MEMS mirrors or optical phased arrays. However, both of these recent approaches lack performance due to the small optical aperture of MEMS mirrors and the low efficiency of phased arrays, according to Lumotive.
Lumotive’s system offers distinct performance advantages, including a combination of:
Large optical aperture (25 x 25 mm) which delivers long range
120-degree field-of-view with high angular resolution
Fast random-access beam steering
Lumotive’s beam-steering technology is the culmination of years of fundamental research into controlling electromagnetic waves using artificially structured metasurfaces. In the past, these concepts have been applied to radio waves, but Lumotive is the first to develop dynamically tunable metamaterials for optics. Their development is a tremendous advance in metamaterials research, as well as a breakthrough technology that addresses pressing and unmet needs in LiDAR and other optical systems.
—David R. Smith, the James B. Duke Distinguished Professor of Electrical and Computer Engineering at Duke University and an Advisor to Lumotive
Lumotive’s LCM chips contain no moving parts and are fabricated using mature semiconductor manufacturing along with liquid crystal display packaging to enable a commercially viable LiDAR system with low cost, high reliability and small size. In addition to cost and performance advantages, Lumotive LCMs can be integrated into small form-factor systems, appealing for other applications in industrial and consumer sectors which can benefit from LiDAR.
Although LiDAR systems will be broadly deployed to enhance perception in diverse markets, including for robots, drones and industrial automation, most observers see LiDAR’s largest near-term application in automotive markets where the technology will enable ADAS as well as fully autonomous driving systems that will first be deployed by self-driving taxi services, commonly called robo-taxis, like those under development by Argo AI, Aurora, May Mobility, Uber and Google’s Waymo subsidiary.
Looks good - better than a rotating system like velodyne, but I wonder how much they cost and will cost ?
You should be able to get better frame rates in the regions where you want them (like in front) which would be good.
When they say Robo-taxi, they usually mean "high cost" in this context, but I suppose you have to start somewhere.
Posted by: mahonj | 21 March 2019 at 02:04 AM
Probably cheaper than mechanical systems, easier to fabricate.
If they're cheap enough, they might be worth making redundant for reliability reasons. Having two sensors with a 60° overlap in their coverage would yield 180° combined FOV and continued operation even if one sensor was failed or obscured.
Posted by: Engineer-Poet | 21 March 2019 at 06:06 AM
Fully redundant and continuous comparative checking-certification, may be what Boeing failed to install on their latest planes.
Driverless vehicles will have to be equipped with fully redundant systems for higher safety.
Posted by: HarveyD | 21 March 2019 at 08:11 AM
temperature control would be a challenge. for auto application the scanner needs to maintain performance over a wide range of temperatures, the LC based steering approach would need a dedicated temperature control unit. and maintaining temperature for a device that big isn't travail
Posted by: Land Buck | 28 March 2019 at 10:19 AM