How Honeywell Made the Leap into Quantum Computing
All signs point to Honeywell being a leader in quantum computing – business president Tony Uttley explains
It sounds like science fiction, but quantum computing is about to change how the world solves complex problems. For the past decade, we have been building technological capabilities and a world-class team to shape how this exciting new industry evolves.
Tony Uttley, the president of Honeywell Quantum Solutions explains how we got here.
What is the origin story of quantum computing at Honeywell?
Almost a decade ago, I was new at Honeywell and leading global strategy and marketing for our controls division. One of my responsibilities included incubating two small businesses. One of them had a simple mandate: Incubate opportunities where there are synergies between Honeywell’s strategy, global macro trends and innovative technologies.
I quickly saw that within Honeywell, we had nearly unmatched capabilities across so many disciplines and technologies, including experts in state-of-the-art hardware and software control systems, advanced electronics, optics and photonics, which largely encompasses lasers, modulators, fiber optics, ultra-high vacuum environments and cryogenics.
At the time, two of our very bright Ph.D. researchers came forward and said, “If we put these technologies together, you know what we could build, right?”
Admittedly, our team did not.
“We could build a quantum computer,” the scientists revealed.
My first reaction was skepticism. I’m an engineer and a science fiction nerd who spent 10 years working for NASA and I thought this was the coolest idea ever, but I also thought, no way. Today, quantum computing is still in its early days, but at that point it was very nascent. We talked about it with our leaders, who were also skeptical but agreed as I did that Honeywell is a technology leader, so let’s make some critical investments and continue to develop these underlying capabilities because we need them in the rest of Honeywell anyway.
A few years ago, when we got to the point where we felt we had a truly differentiated capability, we decided to change our approach. We organized for success, firmly believing that we could be best in the world at this.
Now we have a complete team of technologists, including atomic, molecular and optical physicists, scientists, engineers, and technicians who specialize in the design and development of trapped-ion quantum computers.
So what is quantum computing?
Quantum computing takes advantage of certain properties of quantum physics. Quantum computers exploit the novel properties of qubits (quantum bits) to provide computation power far beyond the capability of classical computing systems. Two properties of qubits give quantum computers their advantage. First, unlike classical computing bits which can only either be on or off, traditionally characterized as being in a “1” state or a “0” state, qubits can simultaneously be in both states at the same time. This is caused by a quantum property called superposition. Creating a collection of qubits, called registers, provides users with an entirely new way of computing. Second, another quantum property called entanglement allows for individual qubits to be linked to each other in such a way that an individual qubit may possess information about the rest of the register. This allows quantum computers to process data simultaneously versus sequentially. These two properties taken together allow developers to compute in entirely new ways that are not available to classical computers and can provide massive speed-ups to run algorithms in record time.
What does this really mean in terms of potential?
While speed improvements over today’s computers are helpful, the real promise of quantum computing is that computations that are impossible for even the best, most advanced supercomputer could be done on a quantum computer because it has that simultaneous computation capability.
Tell us more about how quantum works and what it looks like.
Quantum computers will look nothing like the boxes and laptops of today’s classical computers. Imagine big cryogenic systems that need temperatures that are closing in on absolute zero. Imagine ultra-high vacuum environments, which have five orders of magnitude fewer particles than outer space. Imagine lasers that illuminate miniscule, individual atoms. To give that some perspective, consider that the difference in size of an atom compared to a baseball is the same as a baseball to the entire earth. All of those systems have to be organized to work together in precise harmony. It’s that level of precision, that level of operation that allows something like quantum computation to happen.
What makes Honeywell’s quantum approach different?
In these early days of development in quantum computers, we will have the best, highest fidelity qubits on which to run computations. High-quality qubits are key to being able to develop quantum algorithms. Without quality qubits, identical computations will produce different answers each time. So quality is critical for people who are running computations on quantum computers.
There are multiple ways in which different companies and organizations have formed qubits. While other approaches use manufactured micro-structures to form a qubit, the approach we’ve taken is something called trapped-ion. What trapped-ion gives us is a very high quality qubit. We use atoms of ytterbium (Yb on the periodic table) to operate on within our quantum computer. Because each of these atoms is identical, defined in nature by its atomic structure, our system can be uniformly formed and controlled more easily and quickly compared to alternative systems that do not directly use atoms.
This is why we use the term “Nature’s Qubit” for our system – our qubits are all identical and we can successfully utilize their natural properties. Compared with other systems which have specific, locked locations for qubits, we can move our ions so that we can arbitrarily connect any qubit to any other qubit. This gives you the ability to perform more complex algorithms with fewer computation steps compared with the locked architectures of other systems.
Two reasons. First, Honeywell was already an expert in most of the component parts needed to build and operate a quantum computer. We developed our expertise while solving real problems for various businesses across Honeywell. Our technology development is informed by real-world applications and constraints, and our solutions are fed back in a virtuous cycle.
Second, at its heart, quantum computing is a control problem. And we are the world’s leading controls company. Quantum computing is a natural extension of combining Honeywell’s expertise across multiple technical disciplines.
What did you feel like the first time you saw a trapped ion?
As part of Honeywell’s work on trapped-ion physics supporting an Intelligence Advanced Research Projects Activity (IARPA) program, we first saw a trapped ion in 2014 when working through partnerships. And, as you might imagine, I was awestruck by the fact that we were able to have such precise control over something so incredibly tiny. And as we have quickly replicated our capabilities across six different systems in two different geographies, I was again awestruck by Honeywell’s ability to apply a robust, systems-level approach to such cutting-edge technology.
How do you see something so tiny?
The atom itself is much too small to see, even at the magnification we are doing with the camera. Instead, what we are seeing in images is scattered light (photons) from the atom. Think of it as similar to the astronauts of the space station being able to see light at night coming from cities they are flying over, but are not able to see the individual bulb that is making the light.
Why now and what next?
Global macro trends show a desire for ever more computational power. This is happening while classical computing is showing signs of stalling, where power density and quantum effects are limiting the performance of ever-shrinking transistors. The world is building up this gigantic set of data. The Internet of Things (IoT) and Industrial Internet of Things (IIoT) are providing a rich data set and along with it a need to solve important problems in machine learning. Quantum computing provides the promise of exponential performance improvement to meet these needs.
What industries will quantum help soon?
It’s expected that many industries will benefit from quantum computing. We believe that chemical, materials, and pharmaceuticals will be the first to see benefits, as they can extract meaningful insights from quantum computers with even small numbers of qubits. If a pharmaceutical company can reduce development time, for example, they can potentially save billions of dollars. As qubit count increases and algorithms advance, we expect artificial intelligence and machine learning applications will enable a variety of other industries.
However, it’s early days. Classical computers have been around for six decades or more, and we couldn’t have predicted they would wind up with the capabilities they have today. We expect quantum computers to be able to solve more complex problems as time goes on. The quantum computers of the next few years will provide a test bed on which to develop quantum algorithms. This is the time to look at the algorithms that will make the most of quantum computation.
What is Honeywell offering?
Our objective is to help customers prepare for and capture value from quantum computing. Our experts can already provide much-needed consulting, design of experiment, engineering and development support to companies and organizations that will be profoundly impacted by quantum computing. Before the end of 2019, we expect to have a system that people can access to test their algorithms. We have a testbed for researchers, companies, and physicists to better understand quantum systems and explore technology development and make sure they are going to get the most out of quantum computing.
We believe that quantum computing will eventually have a significant impact on all multi-industry companies. When that happens, those companies will need to figure out how to respond to either take advantage of or mitigate the business changes stemming from the technology. Our approach is to influence how quantum computing evolves and shape the opportunity for Honeywell and our customers.