Why Your GPS Receiver Isn’t Bigger Than a Breadbox

Bradford W. Parkinson shepherded the first GPS constellation to launch and pushed for civilian access By Tekla S. Perry

Photo of Parkinson

Date of birth: 16 February 1935

Birthplace: Madison, Wis.

Education: B.S. in general engineering, U.S. Naval Academy, 1957; M.A. in aeronautics and Astronautics, Massachusetts Institute of Technology, 1961; PhD in aeronautics and Astronautics, Stanford University, 1966

Current position: Professor emeritus, recalled to active duty, Stanford University

First jobs: Supermarket carryout and stock boy, general labourer in construction, newspaper delivery

First tech job: Surveyor for construction projects

Most surprising job assignment: As an instructor at the Air Force Academy, flying 26 air combat missions to troubleshoot the electronics on the AC-130 gunship

Patents: Seven

Most recent books read: The Winter Fortress: The Epic Mission to Sabotage Hitler’s Atomic Bomb, by Neal Bascomb

Favourite book: Tortilla Flat, by John Steinbeck

Favourite music: Classical, particularly Sergei Rachmaninoff, Edvard Grieg, and Ludwig van Beethoven

Favourite food: Spaghetti with meatballs

Favourite restaurant: Café Roma, San Luis Obispo, Calif.

How his spouse would describe him: Intense

After hours/leisure activity: These days, hiking, snowshoeing; in the past, sailing, skiing

Car: BMW Z4 and a GMC truck

Organizational Memberships: IEEE, Institute of Navigation, Royal Institute of Navigation, SME, American Institute of Aeronautics and Astronautics

IEEE Medal of Honor: “For fundamental contributions to and leadership in developing the design and driving the early applications of the Global Positioning System”

Other major awards: ASME Medal, Charles Stark Draper Prize for Engineering, Marconi Prize, Royal Institute of Navigation’s Gold Medal, Institute of Navigation’s Thurlow Award

As I drive through the vineyard-covered hills of San Luis Obispo, Calif., the tiny Global Positioning System receiver in my phone works with Google Maps to alert me to upcoming turns. The app reassures me that I’ll arrive at my destination on time, in spite of a short delay for construction. How different this trip would have been in the pre-GPS era, when the obscured road sign at one intersection would likely have sent me off track. I have a weak sense of direction, and getting lost—or worrying about getting lost—was a stressful part of my life for a long time.

This GPS-guided journey is taking me to Bradford W. Parkinson, the person who made GPS technology—a tool we now take for granted—come together. Parkinson is being awarded the 2018 IEEE Medal of Honor for leading the development of GPS and pushing its early applications. “Just don’t call me the inventor of GPS,” he says moments after we meet. “I was a chief advocate, the chief architect, and a developer, but I was not the inventor.” How about “leader”? “Even that’s overblown. I surrounded myself with guys who would not fail.” Brad Parkinson may be modest about his contributions, but it’s hard to dispute that he was the person who turned a pie-in-the-sky vision of navigating by satellite into a reality.

Parkinson’s preparation for his GPS role began early, with a passion for maps. The walls of Parkinson’s boyhood bedroom were covered with large maps of northern Minnesota’s Boundary Waters—lakes and streams he loved to explore by canoe. “It was easy to get lost,” he recalls. “You had to keep your wits about you.” Then there was his summer job in 1957, just after graduation from the U.S. Naval Academy, as a surveyor on a large construction project.

His graduate education put down another stepping-stone. Sent to MIT in 1960 by the U.S. Air Force, which he’d joined rather than the Navy, he took several courses in inertial navigation. Charles Stark Draper was teaching them, and so it was an irresistible opportunity. That coursework led to a three-year post as chief analyst for inertial navigation systems testing at Holloman Air Force Base. In 1964, he headed off to Stanford for PhD studies. His thesis advisor was Benjamin Lange. “Ben wanted to put a free-rotor gyroscope in orbit to test the general theory of relativity,” Parkinson says.

Parkinson invented a sensor that could tell the position of the rotor relative to the desired axis. Using an algorithm he called hemispheric torquing, he could then apply a magnetic field to adjust the rotor’s position, sending it spinning along the desired axis without changing its overall position in space. Parkinson’s technology is still in use today in some highly accurate inertial navigation systems.

As Parkinson’s knowledge of navigation and space systems grew, the seeds that became GPS were being planted by others. In 1960, the U.S. Navy began testing its Transit program, a satellite-based method of updating the inertial guidance systems used by submarines. Transit’s system worked with as few as four satellites (though the constellation typically included more) in low polar orbits. Along with a network of ground stations, the satellites allowed slow-moving vessels to determine their longitude and latitude a few times a day with an accuracy of about 100 meters.

Ivan Getting, president of the Aerospace Corp., didn’t think that was good enough. In 1962, he started campaigning for a three-dimensional satellite-positioning system that would be more accurate and always available. Getting told me some years ago that he promoted this vision to the presidential science advisor, the heads of the armed forces, and anyone else he thought could have influence, trying “to get the damn thing funded.” Getting’s evangelism led to an Air Force–sponsored study of space-based navigation. The final report, by James Woodford and Hiroshi Nakamura, was published in 1966, although it remained classified until 1979. It laid out 12 main techniques, including the one that became GPS.

In 1972, Parkinson’s path and satellite navigation’s evolution collided. Parkinson had spent the previous year studying at the Naval War College, in Newport, R.I., and sailing whenever he could. Up next would likely be an assignment in the Pentagon. Then he got a call from a colonel who was part of a group known as the Air Force’s inertial guidance mafia. “This wasn’t a black-hat organization,” Parkinson explains, “just people who had gone through the MIT inertial guidance course who looked out for each other.” The colonel recommended that if Parkinson wanted to build systems rather than just analyze them, he should join the Advanced Ballistic Reentry System (ABRES) program, in Los Angeles.

Parkinson, just promoted to colonel himself, took the advice and moved to Los Angeles. He’d been at ABRES a little over three months, working on advanced nose cones and other missile technology, when he was identified as a perfect fit to take over a satellite navigation program called 621B. Parkinson had the right qualifications, but he didn’t want the job. “The consensus was that the program was going nowhere, that it was absolutely dead,” he says.

But it was a three-star general making the offer, Parkinson recalls, and you don’t say no to a general—usually. Parkinson said he’d take the job if he could be named program manager. Anything less, he said, would have been a downward move and wouldn’t have allowed him to control the program—“ and the program was in deep trouble.” The general refused. “So I said, ‘Then I don’t volunteer,’ ” Parkinson recalls. “He wasn’t used to brand-new colonels saying ‘No’ to him.” Parkinson walked out of the office—but he had barely gotten through the door, he says, when the general called him back. Parkinson got his title and took over 621B in mid-1973.

The 621B program aimed to create a satellite-based navigation system that would work almost anywhere in the world. The team had already developed much of the plan and wanted to demonstrate it using four satellites—not an inexpensive proposition. Parkinson began by going through every piece of the proposal with his engineers.

“We were a little worried when he first came on,” recalls Walter Melton, an engineer assigned to the project from the Aerospace Corp. “We heard that he was from the inertial mafia.” The engineers were concerned that Parkinson would be biased against satellite navigation, which was considered a competitor to inertial navigation. “But after the first several weeks it became clear he understood and was a supporter.”

In August 1973, Parkinson presented the proposal to the Defense Systems Acquisitions Review Council at the Pentagon. “I told all these generals and senior civilians sitting around a table what I was trying to do, and then they took a vote,” he recalls. The vote was “No.” Malcolm Currie, then undersecretary of defense research and engineering, chaired that meeting. At the time, Currie was spending a lot of time near his home in the Los Angeles area, preparing to move his family to Washington, D.C. During one of Currie’s Los Angeles trips, Parkinson gave him a one-on-one tutorial on satellite navigation that took up most of an afternoon.

Parkinson now thinks that afternoon was the reason satellite navigation didn’t die after the “No” vote. Indeed, it made an ally of Currie, who quickly reminded Parkinson that the concept presented was merely one he had inherited, not developed himself. Parkinson reported in an oral history that Currie told him, “Listen, you did a very, very nice job, but you and I know that this is not truly a joint program…. Go back, reconstitute it as a joint program, and bring it to me as quickly as you possibly can, and I am very, very certain that we are going to approve it.”

Parkinson and his engineers worked over Labor Day weekend to develop a new architecture for their satellite-navigation system. They met at the Pentagon rather than in L.A., he says, “because too many people associated with the program were entrenched in old ideas.” Gathering in offices that were vacant because of the holiday, Parkinson says, “We hammered out what we wanted to do, and we summarized it in seven pages.”

Parkinson recalls that the “Lonely Halls” meeting, as it came to be known, led to several key changes: The system’s code-division-multiple-access (CDMA) radio signal was modified to include a civilian signal as well as the protected military signal; the orbits of the satellites were adjusted to reduce the number of satellites needed at the optimal altitude, considering the range of available launch vehicles; and the design embraced orbiting atomic clocks, which would free ground-based receivers from the need to keep precise time.

Parkinson says this third change was the most risky—atomic clocks that could handle space radiation did not yet exist. But he knew that Roger Easton at the Naval Research Laboratory was developing a space-qualified atomic clock as part of the Navy’s Timation satellite navigation program—and he bet that some version of that clock would be available for the demonstration satellites.

This decision turned out to be critical for the cheap, small GPS receivers that consumers use today. If instead we all had to carry around superaccurate clocks, the receivers would be vastly more expensive and as large as a stack of dictionaries, Parkinson says. They also would require periodic synchronization to maintain accuracy. They certainly wouldn’t have turned into a tiny package of electronics costing a fraction of a dollar, tucked inside every cellphone.

And Parkinson badly wanted consumers to use the new system. The mission of the project, in his view, was always twofold—extraordinary accuracy and affordability. He even hung a wooden plank above the entrance to the project’s offices in Los Angeles to reinforce the message: “The mission of this office is to drop five bombs in the same hole and to build a cheap set that navigates—and don’t you forget it.”

Parkinson spent the months after the Lonely Halls meeting selling the proposal to Pentagon staff and decision-makers. He flew to Washington as often as twice a week, holding some 60 meetings in two months (he still has a list).

He parried every doubt: Yes, the signal would be powerful enough to be detected in the surrounding noise. Yes, the system’s 10-meter accuracy was achievable. Yes, US $180 million would cover the constellation of four satellites and related ground equipment. (The final price was about $250 million, but that included two added satellites—not a horrible overrun, Parkinson says.)

“He was quite the salesman,” says Melton, his colleague in the 621B program.

The Defense Council approved the proposal in December 1973. Parkinson led the program for three and a half years, until the first GPS satellite was up in space and initial tests verified that the system worked as designed.

Easton’s atomic clock, it turns out, was not ready for that initial launch, but Parkinson had engineers at Rockwell International also working on a space-worthy atomic clock, which was ready. Parkinson still gives Easton credit.

“Easton convinced me that we could do it—and that made a heck of a difference,” he says.

The full 24-satellite system became operational in 1995. The Russian GLONASS system, a similar project begun during the Soviet era, was also completed in 1995. Both the European Union’s Galileo system and China’s BeiDou system are expected to be completed in 2020.

Parkinson retired in 1978 from the Air Force, but he didn’t leave GPS behind. After several positions in industry (including vice president of Rockwell International’s Space Systems Group and vice president at Intermetrics), he returned to Stanford in 1984, this time as a professor of aeronautics and astronautics. He immediately rejoined the orbiting gyroscope project, now called Gravity Probe B, as program manager and a coprincipal investigator; it successfully launched in 2004.

He also led a research group aimed at developing civilian applications for GPS technology. That work led to a robotic sailboat, the first GPS-guided landing of a commercial aircraft, and a system of ground stations that would improve the accuracy of GPS positioning by monitoring and correcting the satellite data. The last project evolved into the Federal Aviation Administration’s Wide Area Augmentation System, which uses data from ground stations to improve GPS’s accuracy by correcting errors in the signal caused, for example, by orbital drift and delays introduced by the atmosphere.

Parkinson’s group also developed the application he is most passionate about today: automated tractors. He had kept automated tractors in his sights for some time; he listed the application as part of GPS’s future in talks he gave as early as 1978.

It wasn’t until around the 1990s, though, that he got his chance to push the technology along. At Stanford, he met with a representative from John Deere who was building ties between the company and universities. Parkinson demonstrated a GPS-guided self-driving golf cart.

“Think tractor,” he told the visitor.

The Deere representative was skeptical that farmers would buy such a system, Parkinson recalls, but the company was eager to partner with Stanford and agreed to fund a development project.

“They sent us about $900,000 and two huge tractors,” Parkinson says. A team of students spent several years developing the technology, first demonstrating a fully functional system in 1997.

Watching the rise of GPS-guided precision farming since then has been gratifying, he adds. Parkinson’s home overlooks a farm, and he often walks in the fields, sometimes spotting, to his delight, a GPS-guided tractor at work. “The tractors pay for themselves in savings in fertilizer and in time.”

“I think he likes the agricultural application because it brings home that GPS is for everyone,” says Penina Axelrad, a former Ph.D. student of Parkinson’s who is now a professor at the University of Colorado. “Now, of course, GPS is in everyone’s smartphones, but that was an early application that everyone could value.”

Parkinson is now mostly retired, though he still has research projects running at Stanford.

He remains one of GPS’s biggest fans—he has more than a dozen devices in his house and car that use GPS, including a watch he wears most of his waking hours.

“He just gets so excited when he sees cool things enabled by GPS,” says Axelrad.

He is also quick to protect GPS when he feels that it’s threatened. Right now, he sees a big threat coming from Ligado Networks, which aims to create a broadband wireless network using the 1,525- to 1,559-megahertz frequency band. This band is adjacent to the frequencies used by GPS, which are between 1,164 and 1,587 MHz, nestled among other bands essentially reserved for satellite communications. Ligado’s band is reserved for Earth-to-satellite communications with some limited use of cell towers to help users connect to the network. Back in 2011, however, the U.S. Federal Communications Commission considered giving Ligado’s predecessor, LightSquared, a conditional waiver to use the frequency band for unlimited ground-based communication. The GPS industry protested, showing data on interference, and in 2012 the waiver was pulled. But Ligado recently came back with a proposal for a lower-power system that it says won’t interfere with GPS.

That proposal is still before the FCC. But testing by the Department of Transportation shows extensive interference, Parkinson says, particularly for the most accurate devices. He’s been working on an editorial to alert the public to the danger of the proposal. Ligado aims to change the designation “of a quiet signal from space to powerful ground transmitters,” he writes. “They would apparently use this to compete with the existing broadband companies. This country already has at least four broadband providers but has only one GPS.”

Says Parkinson, “We endanger it at our peril.”

This article appears in the May 2018 print issue as “GPS’s Navigator in Chief.”

Ultrasound-Powered Nerve Implant Works Deep in Body

An implant that goes in without surgery could help make “electroceuticals” a reality- By Samuel K. Moore

A millimeter-scale programmable nerve stimulator looks like a lumpy rectangle with two metal posts next to a much larger U.S. penny.

Engineers at Stanford University have built a new kind of millimetre-scale nerve-stimulating implant that beats all others of its size at a crucial parameter: how deep inside the body it can operate. The 6.5-millimeter-long programmable implant can receive both power and data via ultrasound through more than 10.5 centimetres of tissue. That’s deep enough for most any application, say its inventors. And because of its versatility and small size—with some modification it could be injected through a needle rather than requiring real surgery—they envision that it will greatly expand the number of conditions treated with electrical stimulation of the body’s nerves.

So far, most of those treatments have focused on stimulating spinal nerves for controlling pain and the vagus nerve for epilepsy and depression. However, researchers have been working on expanding the role of such “electroceutical” treatments to include ending postpartum bleeding, alleviating rheumatoid arthritis, and restoring bladder control, among many others. Because implants today require surgery, “implantable devices are seen as a last resort solution,” says Stanford University assistant professor electrical engineering Amin Arbabian. “If you have a disease with any other solution you’ll probably opt for that.” But a nerve stimulator that can be implanted with minimally invasive surgery or simply is injected would allow nerve stimulation treatments to reach 100-fold more patients, he argues.

The Stanford implant consists of: a piezoelectric receiver that converts ultrasound applied from outside the body to electricity, a capacitor for storing that electricity, two stimulating electrodes, an LED, and a custom chip to control it all. Those components are all inside a biocompatible package about the size of a fat grain of rice or a rather slim tic-tac.

The device is capable of extraordinary flexibility in its electrical stimulation parameters. It’s programmable via data sent through the ultrasound signal, allowing the stimulation’s amplitude, pulse width, and frequency to be adjusted to whatever recipe will best interface with the body’s peripheral nerves. “The reason we’re able to do all this is that we have orders of magnitude more power available at large tissue depths than in conventional wireless implants,” says Arbabian.

Getting power and data deep inside the body has always been a problem, but it gets more difficult the smaller the implant is. “Our aim was to design it to be very small and operate very deep,” says Jayant Charthad, one of Arabian students. “This is the hardest problem for transferring wireless power and data.” Comparing the ratios of tissue depth to system volume, the Stanford stimulator beats other experimental systems by at least an order of magnitude. “This has been a long-standing challenge for developing electroceuticals,” he says.

Arabian’s team chose ultrasound to carry the power and data rather than radio signals as other implants do, because its small wavelength matched the millimetre size of the implant and because ultrasound can penetrate far into the body without harming intervening tissue. But even so, the system required considerable design work so that the implant could make the most of ultrasonic energy. “It’s taken five or six years and five or six students,” says Arbabian. “It’s not a straightforward application.”

Now that they’ve built working stimulators, it’s time to test them in living things. One early application is likely to be as the stimulator in a bladder control system, Arbabian says. But first, they will be conducting tests in animals. That’s where the system’s LED will come in handy. At the same time it produces electrical stimulation, it can provide the light for so-called optogenetics experiments—where light triggers the actions of nearby genetically modified neurons. Being able to coordinate and compare electrical and optical stimulation “would open up a lot of science,” says Arbabian. “We can use this as a platform for discovery.”

How to Bend Diamonds

Flexible nanodiamonds promise to open up novel optical and electronic properties-By Dexter Johnson

“Experiment (left) and simulation (right) of a diamond nanoneedle being bent by the side surface of a diamond tip, showing ultralarge and reversible elastic deformation.”

Yang Lu, Amit Banerjee, Daniel Bernoulli, Hongti Zhang, Ming Dao and Subra Suresh

It’s commonly known that diamonds are the hardest natural material. However, with that hardness comes brittleness: they may be hard but they’re not very flexible.

Now an international team of researchers has demonstrated that diamonds, which are commonly believed to be inflexible, can be bent and stretched significantly. The researchers showed that the maximum tensile elastic strain of a diamond can reach nearly 9 percent, close to the theoretical limit of the material.

The researchers believe that these enhanced mechanical properties make nanodiamonds much more durable than expected, and therefore could lead to applications that involve mechanical loading, making them candidates for applications such as diamond needle-based intracellular delivery. But it is what this flexibility does to diamonds’ optical and electrical properties may prove to be the most significant in the long run.

In research described in the journal Science, scientists from the City University of Hong Kong, China, the Massachusetts Institute of Technology,Ulsan National Institute of Science and Technology (UNIST) in South Korea and Nanyang Technological University in Singapore produced nanoscale diameter (about 100-300nm) diamond needles using reactive ion etching (RIE) of CVD (chemical vapor deposition)-grown diamond thin films.

The resulting material demonstrates that the mechanical properties of even hard, brittle crystalline materials like diamonds can be fundamentally changed when their sizes are reduced to nanometer-length scale.

While the purity of the internal crystalline structure and surface smoothness of the material is important, the key to achieving such flexibility in diamonds is actually the “size”—the dimension/diameter of the nanodiamond needles, according to Yang Lu, an associate professor at the City University of Hong Kong.

Flexibility on its own is great for using diamonds in applications that require more mechanical loading, but there is also a quantum mechanical effect that comes with this bending.

SEM image sequence of bending deformation of a typical polycrystalline nanoneedle, where (B) shows the maximum deformation before fracture and (C) shows the nanoneedle immediately after fracture has occurred.

Images: Yang Lu, Amit Banerjee, Daniel Bernoulli, Hong Zhang, Ming Dao and Subra Suresh. SEM image sequence of bending deformation of a typical polycrystalline nanoneedle, where (B) shows the maximum deformation before fracture and (C) shows the nanoneedle immediately after the fracture has occurred.

Previous theoretical studies showed that when elastic strains exceed 1%, quantum mechanical calculations indicate significant physical and/or chemical material property alterations due to the changes in energy band gap structures, according to Ming Dao, principal investigator and director at MIT’s Nanomechanics Lab, who was a co-author of the research.

“These property alterations may include significant changes in mechanical, thermal, optical, magnetic, electrical, electronic and chemical reaction properties, and could be used to design advanced materials for various applications through the emerging ‘elastic strain engineering’,” said Dao. “When maximum elastic strains can be changed in real-time between 0 to 9% in nanodiamonds, there is a lot of potential for exploring unprecedented material properties.” After two years of careful iterations between simulations and real-time experiments, Dao said that he and his colleagues have managed to streamline a nanomechanical process that can precisely control and quantify the maximum amount of elastic deformation within the nanodiamonds.  The resulting method enables accurate control and on-the-fly alterations of the maximum strain in the nanoneedle below its fracture limit. This also means, of course, that its electronic properties can be changed on-the-fly as well.

While this level of elasticity in the nanodiamonds can change their band gap structures, incorporating impurities into the severely strained lattice of the nanodiamonds may lead to revolutionary changes in diamond’s electronic and optical properties, according to Yu. “This could open up a lot of novel optoelectronics applications for diamonds, such as more powerful or colourful laser or maser (microwave amplification by stimulated emission of radiation),” said Yu. The first step toward commercialization of these applications will require researchers to microfabricate diamond nanostructures in well-defined geometries and crystalline structures.

For optoelectronics applications, another challenge is to quantify and control the local optical/electronic property changes, in real time, for a single diamond nanostructure under elastic straining, according to Lu.

Lu added: “We will work with physicists and electrical engineers to explore the optoelectronics applications of the elastic nanodiamond structures, and we may even find its applications in the emerging diamond-based quantum computing technologies.”

 

Nanoscale 3D Printing Technique Uses Micro-Pyramids to Build Better Biochips

This nano printing process allows researchers to 3D print more material on a biochip than ever before, making it easier to study biomedical issues- By Christina Dabney

https://spectrum.ieee.org/video/biomedical/devices/nanoscale-3d-printing-technique-uses-photochemistry-and-micropyramids-to-build-better-biochips

Making biochips, a key technology in studying disease, just got a little easier. This new nano printing process uses gold-plated pyramids, an LED light, and photochemical reactions to print more organic material on the surface of one single biochip than ever before.

The technique uses an array of polymer pyramids that are covered in gold and mounted on an atomic force microscope. These arrays, which are one square centimetre in size, contain thousands of tiny pyramids with holes that allow light through and make sure that the light goes only to specific places on the surface of a chip below, immobilizing delicate organic reagents on the chip’s surface without damaging them.

Processes like this, known as tip-based lithography, are widely considered to be the best way to 3D print organic material with nanoscale feature resolution. But in the past, they were limited by the fact that they could only print one kind of molecule at a time. Now researchers at Hunter College and the Advanced Science Research Center (ASRC) at The Graduate Center of the City University of New York think they have solved that problem.

They’re using microfluidics, the manipulation of fluids on a molecular level, to expose each biochip to the desired combination of chemicals. Then, they use photochemistry to shine light through the apertures in the pyramids. As the light reacts with the molecules, it adheres them to the chip. With typical tip-based lithography systems, the light can overpower the chip, destroying some molecules. But the CUNY research team uses beam-pen lithography, where the light is confined and channelled through small apexes. This allows the team to control the light and protect the organic materials that they have already printed on the biochip.

Adam Braunschweig, the lead researcher and an associate professor with the ASRC’s Nanoscience Initiative and Hunter College’s Department of Chemistry, says this method of 3D printing biochips will help scientists understand cells and biological pathways. That’s because this technology should make it easier and more efficient to study disease development and solve other biological puzzles, such as detecting bioterrorism agents.

BMW Will Use Solid-State Lidar From Innoviz

It’s the first publicly acknowledged deal between a major car company and a solid-state lidar firm- By Philip E. Ross

Still image from Innoviz showing Innoviz LiDAR Based SLAM.

Israel-based Innoviz has announced that it will supply solid-state lidar to BMW. The device, along with radar and other systems, will be incorporated into a self-driving package from Magna, a major auto supplier.

Innoviz says that when volume production begins, the lidar’s price should drop to the hundreds of dollars, down from the “single-digit thousands” that today’s test units go for. The company says that it can now make several thousand units a month on its existing assembly line, in Israel, and that it’s building another line in China.

The company argues that today’s deal with BMW vindicates the solid-state approach to lidar, in which the laser beam is steered without machinery. Innoviz does the trick with microscopic, moveable mirrors. Most recent lidar startups also use solid-state approaches. But Velodyne, the industry leader, rotates its lasers on a roof-mounted tower, and the startup Luminar uses a macroscopic mirror to sweep its beam around. Luminar recently announced the start of mass production, with some of its units going to Toyota’s self-driving research team, and the rest to other, yet-unnamed automakers.

Image of Innoviz's LiDAR

In its release, Innoviz said only that its current test unit offers “unrivalled angular resolution at the highest frame rate of any LiDAR solution currently on the market.” But earlier this year, a company spokesman told that its test units had a frame rate of 20 frames per second, an angular resolution of 0.15 by 0.3 degrees, a detection range of up to 150 meters, and a field of view of 73 by 20 degrees.

By contrast, Luminar says its mechanical system can vary the frame rate from as slow as one per second, for high resolution, to as fast as 20 per second, in exchange for lower resolution. It says its lidar can detect objects 250 meters ahead. Velodyne claims a 300-meter range for its top-of-the-line system.

It’s hard to say which design will win, though. Different self-driving packages give different weights to the various sensors; one might rely on the radar rather than lidar to cover the longer distances. Of course, another very important factor is that of cost. And solid-state systems generally cost the least.