Transcript | Quantum Navigation and other Aviation Use Cases with Boeing

Boeing is all about connecting the world and soaring to new heights in every sense of the phrase. Nestled inside this giant is a large team dedicated to using quantum computing and sensing to ensure innovation in aeronautics, ranging from materials science to navigation and other use cases. Join Host Konstantinos Karagiannis for an uplifting chat with Jay Lowell from Boeing.

Konstantinos

Boeing is all about connecting the world and soaring to new heights, in every sense of the phrase. Nestled inside this giant is a large team working on using quantum computing for materials science and quantum sensing for navigation. Find out just how soon quantum applications will be up in the air in this episode of The Post-Quantum World.

I’m your host, Konstantinos Karagiannis. I lead Quantum Computing Services at Protiviti, where we’re helping companies prepare for the benefits and threats of this exploding field. I hope you’ll join each episode as we explore the technology and business impacts of this post-quantum era.

Our guest today is the chief scientist for disruptive computing and networks at Boeing, Jay Lowell. Welcome to the show.

 

Jay

Thanks for having me.

 

Konstantinos

It’s great to finally get you on here. We’ve known each other a little bit — we’ve met at several CQE events and things — and Boeing is quite impressive in the quantum space. For quite a few years now, you’ve been generating some news. I wanted to talk about all the exciting things going on at the company, but first, give our listeners a little background on how you got interested in quantum computing — I know your background is in physics, but how you got here.

 

Jay

Let me give you the abbreviated version. I am trained as a physicist. I was in the military. I went to the Air Force Academy and studied to be a physicist in the air force but ended my career at DARPA as a program manager. And while I was at DARPA, I had a number of programs in quantum technology space and worked with some of the original quantum computing and quantum information science research programs that DARPA had. Some of my programs that I ended up funding while there were developing, for instance, quantum sensors for navigation, some methods of doing quantum nonlinear optics that are turning out to be useful in some of these quantum networking protocols and domains, and a program to do quantum simulation that ended up producing a lot of the underlying technologies that are used by neutral-atom and ion-trapped quantum computing.

A lot of the control systems and laser control mechanisms, a lot of that technology was developed in that program, which set out the ambitious goal to build a quantum simulator to simulate quantum materials using either atoms or ions held in lattices that you could do an analog control of that system, create a Hamiltonian, make it work like an analog to the quantum material you’re trying to model and search for novel quantum materials. It was an ambitious program and was a nice way to walk out and end my career at DARPA.

Then came what I call my seven years in the desert, where I walked away from that field. I went and worked in software. I came to work for Boeing, and I worked in the commercial airplane side of the company more than I worked in this quantum technology space. I did work on a number of things — battery safety, electromagnetic effects — how we protect our airplanes against lightning strikes — and factory automation and technologies for building and assembling our airplanes.

But around the end of 2018, beginning of 2019, my current boss and the CEO of the company were looking at ways for the company to place bets on long-term technologies that we wanted to develop, and quantum technologies is one of those that came up. And I was one of the few people in the company who had a background and expertise in not just that technology area, but also in developing technologies relevant to that area, and so he brought me in as the chief engineer for this organization, and that started what’s now a three-and-a-half-year ride in building up the capabilities of the company and starting our journey to develop quantum technologies that we feel will help our company move into the future.

Like you’ve said, it’s been a long, winding road, and an interesting journey.

 

Konstantinos

What’s interesting is, it sounds like everything you did at DARPA is cut-and-paste, drop it onto Boeing several years later, and you’re doing that same general theme again, where you’re developing materials. Then, of course, the quantum sense that we’ll get to today. It sounds like you’re covering both areas.

 

Jay

Similar. There are a lot of similarities, but there are very distinct differences as well. Some of the things that we’re doing at Boeing are capitalizing on capabilities that we have within the company that are unique — or at least we feel they’re unique — and we feel that they give us an opportunity to get ahead of the field and make a contribution by leveraging that capability.

The other thing that’s unique is that we have a whole host of efforts that I would say are inwardly focused. You’ve heard me talk about our quantum computing applications team. That team is focused on how we internally are going to be able to use the quantum computers as they develop, and understanding how to align that to the mechanics of how our company works, how we engineer our products, how we develop materials and qualify those materials, how we model and simulate our products to understand their performance and understand the design decisions we have to do. Those are the things that we’ve been addressing in our compute applications team.

It’s very different than the approach that I took at DARPA, which is, if it’s generally applicable to developing the advance in the field, that’s something DARPA was willing to invest in, and here we have focused on activities that are going to either help us develop technologies that we see have a use in our business or help us make our business more effective, more efficient and better. The areas are similar, but why we’re in those areas is very different, and it results in very different projects that we are funding and very different activities that we end up doing.

 

Konstantinos

In the quantum industry, there’s usually a lot of specialized companies that deal with some aspect of the stack, and then there are customers that need help with a specific problem, and then, every once in a while, there are leaders in their field that take on quantum, and they end up with a workforce that’s rather large and a focus that’s rather large. JPMC comes to mind, for example. You can almost say that they’re the biggest quantum company hidden inside of their company. Not hidden — it’s in the open. How does that look at Boeing? It sounds like you have a pretty decent-sized team. What kind of categories of staff do you have dedicated to quantum?

 

Jay

It’s hard to say, precisely, because it changes month to month. We’re organized around the portfolio projects, and the complicated reason is, because we’re organized around a bunch of projects, we actually don’t own any of the people. They don’t have a reporting relationship to our organization. We’re a project organization.

We essentially define the projects and have the resources to make those projects work and manage those projects, but then we go out and ask for people from our engineering and research organizations, who come in and do the work for those projects. It fluctuates, because as the needs of the project change, we’ll have people who work for a period of time on a particular piece of a project and then go off and start doing something else. But the general answer is, we’re steady-state funding thirtyish people — maybe sometimes a few more, sometimes a few less. There are more than 30, but 30 is a number I’m comfortable saying.

I have a couple of projects for which I’m not yet able to talk about what we’re doing, and I can’t say how big those projects are, because I’d give away information that we’re not yet ready to share, but I’ll give you an example. We had a review of all of my projects for the first time — an in-person review — last week, and we had about twenty-some people come from all over the country and sit down and go through a two-day project review where we reviewed all of the projects and activities over the past year, what their plans are for the next year, and what their technical challenges are, what their milestones are that they’re supposed to meet, and then looked for opportunities for collaboration across projects — where are places that project A can help project B?

Each project lives in its own little world, but we have to take times like that to have everybody take a step, pause, look up, look around and see what’s going on around them, and we’ve found a number of opportunities for projects to help each other go faster by taking a little bit of time to share, “We did this, and we solved this problem over here, and it seems like it’s related to your problem that you’re trying to do over here. Maybe we can help each other figure these aspects out,” and we’re able to do that, and even though I see all these projects, I don’t see them at the level of detail all the time that’s able to find those opportunities and those little nuggets of collaboration that takes bringing everybody together and forcing interaction, if you will. It was pleasant to see — let’s put it that way.

 

Konstantinos

It sounds like a great environment for people who are looking to expand their skills. A lot of our listeners want to know how they can get involved in quantum, and so, with different guests, I ask them different paths that they see forward for, let’s say, becoming a quantum coder or whatever. It sounds like you have some kind of a system in place for pulling people in and then getting exposure to something that’s new to them in that way.

 

Jay

Yes, we’ve been doing that systematically, again, over the last three years. When we started out the organization, we had five people. We’re now at 30. We’ve figured out a way to grow by finding people with requisite skills within the company, by hiring people in who have skills that we need. In terms of the people who have a straight quantum background, it’s not the majority of the people who are doing work. More than half of the people have a nonquantum background working on our projects.

I feel like it’s one of the reasons many of the projects are successful. For example, a lot of times, you talk to a small company and they’ll have physicists doing everything. One of the projects I wanted to highlight is our quantum navigation project. We’re going to be demonstrating a fully quantum inertial navigation system later this year. The schedule keeps slipping by a week, so I don’t want to throw out a specific date, but we’re looking at sometime in early June right now, and we’re excited about this project, but most of the people working on that are navigation engineers. They’re not quantum physicists.

We have one person who’s got a quantum background, and a bunch of navigation engineers, and that has allowed our team to be very successful, because we’re focused on making sure we understand how to use that technology in the context of the application that we need to use it for, and that means we need domain experts from both sides. We need people who understand the quantum aspects of the system, but we need the people who understand how a navigation system is used, assembled, put together, tested, all of those aspects. It’s something that a quantum person could learn, but it’s better to take people who do this every day, and do it much more efficiently, so I can get to make much faster progress by staffing that team that way than I can by hiring a bunch of physicists and trying to get them to learn everything that they need to make the project successful.

 

Konstantinos

That makes sense, and this seems like a good point to start diving into the main areas of use cases that you’re working on there. We could start with the navigation. To me, it seems like a very practical application of quantum sensing.

 

Jay

It is a very practical application, and fundamentally, it is an application that is critical to the business of the Boeing Company. We build things that move. That is the mission of our company — to connect, protect and inspire the people of the world, and in order to connect the people of the world, you have to move people from place A to place B, and so those platforms that we build that move things around and connect those people need to know where they are. Fundamentally, navigation is a key enabler for those platforms to be effective, so that technology is important. And as we build those platforms, we need to make them work in a way that’s safe and reliable, and extremely effective, and so these quantum navigation systems enable our platforms to navigate in times when GPS is not available or when GPS is degraded without loss of capability, and that’s something that has tremendous value to our platforms.

It adds a lot of value to those platforms. Therefore, it’s an area that the company is fully invested in us being good at, and so it makes sense. It’s a use case that aligns fundamentally with our business and forces us to focus on what aspect of the technology we need to develop and how we need to develop it to make it useful for our business. This, to me, is the key to having a good quantum group within a large company — making sure you understand you’re not developing technology because the technology is cool — but it is. You’re not developing a technology because it’s the new thing. You’re developing a technology because you understand how your business needs that technology to work and function in a specific way to make your business better.

 

Konstantinos

Obviously, we can’t get too into the weeds, even just for intellectual-property reasons, but you said “when GPS isn’t available.” This is some precise time measurement that’s happening?

 

Jay

This isn’t about measuring time. This is about measuring forces and rotations. If you go back to your first-year physics or your high school physics class, you learn kinematics — how a ball looms when you fire it through the air, and how you can use equations and motion to predict where that ball is going to be if you know what it’s been doing up to this point, and in the end, that’s what a quantum inertial navigation system does. It uses aspects of quantum sensing to measure the motion — in this case, the motion of atoms — very precisely, and in measuring the motion of those atoms precisely, you’re able to understand how your box that contains those atoms is moving through space.

 

Konstantinos

Like a quantum gyroscope.

 

Jay

It is a quantum accelerometer and a quantum gyroscope. It’s multiple accelerometers and gyroscopes, but that’s the key. Now, we have other sensors that we’re going to be fielding as well, but we’re working hard to put other quantum sensors on there that work differently that’ll allow us to identify where we are by localizing our position on a map, looking at a map of fields — for example, in this case, magnetic fields — and being able to understand where we sit in that area of the magnetic field. We’re going to be using all these quantum sensors together to understand, how good can we make a fully quantum navigation system? And given that, how are we going to turn that demonstration into a product that we integrate onto one of our Boeing platforms?

 

Konstantinos

Does everything remain quantum inside that little box, and then there’s some kind of ASIC that converts it to purely classical information to come out as an output?

 

Jay

Yes. The volume of the system that is quantum is relatively small, and in all the cases of the sensors that we are fielding, they’re using a laser to make the measurement of the quantum system. You’re reading the result of that laser probing, that quantum system, that set of atoms, and from reading that out, you then have a processor that is doing some calculations and converting that into the thing that you want to measure — in this case, either a magnetic field or an acceleration, or a rotation.

 

Konstantinos

That makes sense, because it seems very difficult to implement a quantum network on an airplane to keep all the information in that state.

 

Jay

These sensors are not going to be networked in that sense. There’s obviously a classical communication network, but that is there to essentially take those readouts and report those all back to a central processor, which takes the net sum of all of those measurements and assembles those into a full picture of where that system is in space at all points in time.

 

Konstantinos

Is the hope to get these kinds of sensors to improve other measurements that go on on a plane — pressure, altitude, anything like that?

 

Jay

We have a couple of ways that we hope to use them, but the key thing here is, every airplane already has inertial measurements on it — a set of accelerometers and gyros that help determine the airplane’s orientation and accelerations. We think that by replacing that with a quantum unit, we could provide advantages to that platform, because it’s much more accurate. Therefore, the entirety of the navigation system could handle outages by GPS much more elegantly, much more gracefully.

We’re not advocating building a navigation system that doesn’t actually use GPS. It’s foolish to think about building a navigation system that doesn’t take advantage of enough resources that you have available or enough measurements you could make to make that right balance between the performance of the system and the cost of implementing that system in terms of the dollar cost, the volume of the system. The size, the weight and the power is what typically is laid out. Within the constraints that we have for the size of that system, we want to maximize the performance that we could get out of it by using the best possible sensors we have available.

 

Konstantinos

That’s exciting to have something so real-world for quantum sensing finally. That’s terrific.

 

Jay

We’ve been working on this project now for a couple of years. It’s been in the works for a while. We’ve been going at it quietly until we were confident things are coming together in such a way that we’re going to be able to conduct flight tests. That has been our long-standing goal, which is why we’ve just recently announced that we’re going to do this. It helps me put that last bit of pressure on the team that they’ve got to deliver, and yet they still keep pushing back, saying that we’re going to have to delay the schedule here and there, but that’s the job of the team. The job of the team is to tell me when they’re going to be ready to do it. My job is to tell them I’d like them to be ready earlier.

 

Konstantinos

I’m already cringing at the thought of some reporter writing a story about this and saying that it’s going to take you to a different quantum realm or something when you fly the plane.

 

Jay

I hope that that’s not the case. That’s not the stuff that we’re interested in. This lines up very well with what we’re interested in at Boeing. We talked about our compute applications team as another use case. I think of it as a general use case where we’re looking at many individual use cases for quantum computers.

There are two other areas that we’ve been looking hard at. One is in the quantum networking space. Again, the easy way to think about why this is an important use case for us comes back to that fundamental statement: As a company, we build things that connect people, and we can do that two ways: These platforms move — they can carry people — but we can also connect them together with networks as a means to ensure that we’re bringing people together with information that they need to have.

We have, for instance, satellites that help build global communications networks or global sensing networks, and so the intriguing question about quantum networks is how they are going to help things that work that way, how they are going to help make those network systems better, how we’re going to be able to put into play quantum sensors that are entangled and basically collaborate to make those platforms much more effective at the job that we’re trying to get them to do.

We view it as a long-term research area, but we’ve tried to lay out research projects that help answer what we see as fundamental questions we need to understand to be able to make a more focused set of decisions in the future about how we would go construct, what kind of network we would want to go do. We need to ask the questions now and put the research in place now that give us the ability to make that decision a year or two from now.

 

Konstantinos

And with quantum networks always comes the opportunity to connect quantum computers, and sensors to quantum computers, in all these other locations that open up.

 

Jay

Right. We’re looking at both, but we’re interested in understanding, what are the limitations and constraints around building networks, and so we’re trying to put our projects in place to address those aspects. We’ll come into whether we want to go do the sensing or computing networks later. Once we understand the constraints of the networks themselves, that will help us make a decision about which kind of networks we want to build.

 

Konstantinos

As far as using quantum computing, I know since around 2020, you’ve been working with IBM Systems to work on materials science. Do you want to talk about that — designing materials for planes, and those kinds of applications?

 

Jay

We have been looking at materials science. We’ve done three studies in materials science that are in various stages of maturity. We started off looking at what turns out to be an optimization problem, but it’s an optimization problem of how to optimize aspects of a material.

The composite materials in our airplane wings have stacks of carbon fiber that are oriented in different ways, and that is to help make sure that the structural properties of that material are what we want. It turns out you can’t just randomly lay them up. They have to obey some rules in order to guarantee that those materials will have the properties we want. We looked at using a quantum computer to solve that constraint optimization problem, how to find an optimal structure that satisfied all of the rules, and we had to solve a very large graph optimization problem to do this. We’re very proud of the result. We collaborated with IBM on that project. I think the preprint is still out, where it’s hung up in final review for publication in the journal, but we hope it’s coming any day now.

We’ve also looked at the quantum chemistry problem of corrosion, another big materials science problem that we have, given that we build these platforms that go into inhospitable environments. The metals on the composed parts of these planes and submarines and satellites, they corrode over time. Right now, a lot of that research into corrosion is fundamentally done heuristically by experiments and by building heuristic models that describe the high-level behavior of what this results in.

We want to be able to have an understanding that connects all the way down to the molecular and atomic level and all the way up to those high-level material behaviors. We’ve been trying to understand how to build chemistry models that have those steps and use quantum computers to solve portions of those chemistry models. That’s been a very ambitious project. There are many stages of the calculation that need to be tightly linked in order to make this work. A lot of the work there has been understanding how to lay out the flow of calculations and how they pass information from one to the other to the next — and making sure we understand how to go do that and how to build the process, and how to build some tools to help us understand that process better and think about how to tweak and optimize that to the extent that we can.

Then, more recently, we’ve been looking at damage to coatings from the ultraviolet invisible-light exposure, looking at the underlying molecular mechanisms that we expect cause this damage and trying to figure out if there are ways we can use quantum computers to simulate those behaviors — quantum as well as classical methods — and trying to investigate these damage mechanisms and understand how model these things, and how we can use those models to learn how to devise and design better molecules to do that job that don’t have these side effects.

 

Konstantinos

It sounds like you’re already going to have some advantage from quantum sensing with that box we discussed. With these three general use cases for quantum computing, what does your gut tell you might be the first that would give you some quantum computing advantage?

 

Jay

Well, we can see the point where the optimization problem starts to be beyond the scale of the HPC resources we have in-house. In the community sense, it won’t yet be at quantum advantage, but at that point, it’s a practical advantage for us, because getting time on a faster supercomputer than what we have access to is in itself a difficulty, and if we have access to a quantum computer, which we do through relationships that we have with IBM and now some other companies as well, if that’s faster, then it’s faster, and so, in a practical sense, I don’t care.

 

Konstantinos

That practical advantage — that’s what we call customer advantage.

 

Jay

It’s very much a practical advantage. In the end, the point of those calculations is to be able to keep up with other parts of the design process that are going on simultaneously, and those simulations need to be fast enough that they can keep pace with all of that work, and right now, they’re not able. We are able to do things that are a little bit smaller than what we want in the classical sense. Those calculations take longer than we would like to keep pace.

The quantum computing results, we’re not able to do a system that’s as large as we’re able to do classically right now, but based on our understanding of how these things scale and some investigation of that, we see a point where it crosses and where it gets to the place where it’s now keeping up and maybe even outpacing some of the other calculations.

And now we have a different challenge. Now, we have to work on and understand how we rejigger that engineering work, so that everything is keeping pace together, by working on the next-slowest thing. It’s a never-ending bed of nails: You hammer one down, and you look and, there’s another one. I’ve got to knock that one down, and you have to keep that going to wring all those inefficiencies out of a long, detailed engineering-design process for these platforms and these products.

These things take years to build and design, and these processes have to work — and have to work reliably. A lot of our focus has been on understanding and guaranteeing the outcome, and understanding when we need to make a decision about changing how we perform that calculation without loss of that guarantee.

 

Konstantinos

Anyone who’s flown on the 787 Dreamliner knows what it is to watch the wings bend up in the sky and everything. That’s not subtle.

 

Jay

Not subtle at all. Every time I sit on the Dreamliner, I look at those wings and a warm feeling comes over me, but every bit as warm is the fact that I know that inventions that I’ve worked on are somewhere else on that airplane, helping that airplane fly, because of years of effort trying to devise a better way to do something that’s better, faster, cheaper, safer or a combination of all those things.

 

Konstantinos

It sounds like in the case of this experiment, the next iteration of QPU — let’s say, the next IBM chip, if that’s the one — that might be the one that crosses that line.

 

Jay

It’s probably maybe the next year, but we’ll see. We’re very much a “Do the work and let the work speak for itself” kind of group, which is one of the things that I like about our group — we’re trying to be systematic and data-oriented about this. We have a clear vision as a team, and a relatively uniform approach to how all of those projects approach doing the work that they’re doing. That’s been satisfying — to see the team grow, and to go from just starting to understand what to go do to building a culture that allows them to work in an effective way to starting to come back and report great accomplishments and an even more exciting future. It has been satisfying to see over the past three years. As a company, we’re setting ourselves up for more growth in this space.

We’re still hiring people. We’re going to be hiring people over the next couple of years, we expect. Not superfast, but we’ll be hiring people, and that’s great to see. It’s because we have an understanding of how this is going to be useful. We see what we’re going to have to go do to make that work, and that work requires us to be more aggressive and more accomplished over time.

 

Konstantinos

And I saw that Boeing donated $5 million to UCLA for quantum science and technology. It’s exciting to see that — not only you as a success story in making quantum reality at Boeing, but also you furthering that path of development for people coming up the ranks.

 

Jay

As I’ve said, it’s satisfying that the leadership of the company is committed to the technology. They understand it’s one of those things that we need to have out in the future, and they understand that it’s a long-term play. We’re not going to see return this year. We’re not going to see return next year. We’re going to see things that help give us confidence that we’re on the right path, and we’ve gotten some contract research out of the investments that we’ve made internally, and that’s good to see.

We might be able to get that, but we’re in this for the long game. Boeing’s a company that thinks in 30-year product life cycles. So, we’re not scared by some investment or some decision having a lifetime that’s more than just around the corner or the next quarterly report. We think in those long time frames, and in so doing, that helps us see that strategy, see what we need to do to make sure that that strategy emerges in a way that helps us, and understand how to make decisions along the way that help reinforce that behavior and help position us to be doing what we want to do and make the company a better company and make the world a better place.

 

Konstantinos

With that, I’ll thank you so much for sharing these insights into what you’re doing there. I was looking forward to having you come on, so I appreciate it, and good luck with all these projects going forward.

 

Jay

Thank you. I appreciate the opportunity to speak with you today. It’s always a pleasure to see you, and sorry we couldn’t do this in-person. That might’ve been even more fun, but I’m sure we’ll see each other in the conference.

 

Konstantinos

Yes, I’m sure I’ll see you in person. Thank you.

 

Jay

Thanks.

 

Konstantinos

Now, it’s time for Coherence, the quantum executive summary, where I take a moment to highlight some of the business impacts we’ve discussed today in case things got too nerdy at times. Let’s recap.

With about 30 dedicated quantum staff and the ability to rotate additional experts on a per project basis, Boeing contains one of the larger QIS teams in the industry. Combined with their dedication to education, Boeing sounds like a fertile breeding ground for quantum innovation, and they’re even hiring.

What is this team bringing to the skies? For starters, they’ve developed a practical use for quantum sensing. When GPS is unavailable or degraded, another device, based on inertial navigation, helps pilots know their plane’s location in real time. Quantum sensing will enable a more accurate version of this device with accelerometers and gyroscopes that measure the motion of atoms. Quantum networking will also connect the world, and Boeing’s exploring how to network and deploy quantum sensors in quantum computers. They’re trying to understand the constraints in these early days before they roll out actual quantum networks.

On the quantum computing side, Boeing has been working on material designs for about three years. It’s a different kind of optimization problem that looks to optimize how composite materials are laid out to create structures that satisfy needs, such as flexibility or strength. Corrosion is also a significant concern for planes expected to fly for many years or decades, and quantum computers may help solve some parts of chemistry problems in calculations involved in preventing corrosion. There may also be ways to simulate potential future damage to coatings and materials using quantum computers.

Boeing has a terrific view of quantum advantage, considering practical advantage rather than benchmark advantage. If a step in the process becomes fast enough on quantum to show an edge overall, it’s a win. They know it’s a long game, and they are no strangers to products with multidecade time frames and life cycles.

That does it for this episode. Thanks to Jay Lowell for discussing the exciting projects underway at Boeing, and thank you for listening. If you enjoyed the show, please subscribe to Protiviti’s The Post-Quantum World, and leave a review to help others find us. Be sure to follow me on all socials @KonstantHacker. You’ll find links there to what we’re doing in Quantum Computing Services at Protiviti. You can also DM me questions or suggestions for what you’d like to hear on the show. For more information on our quantum services, check out protiviti.com, or follow ProtivitiTech on Twitter and LinkedIn. Until next time, be kind, and stay quantum-curious.

Loading...