Honeywell is best known for the iconic round thermostat you’ve probably got on a wall in your house or apartment. The company has since stopped making thermostats, even smart ones, leaning instead on aerospace, defense, materials science, chemicals, and more — all crucial elements for building a quantum computer, it turns out. And for the past decade, that’s exactly what Honeywell Quantum Solutions has been quietly working on.
Quantum computers are generally used to solve problems involving massive amounts of data more easily, like optimizing the routes airplanes fly. As Tony Uttley, the president of Honeywell Quantum Solutions, told Digital Trends, quantum computers have “magical-type qualities around them” because of the speed at which these complex problems can be solved.
Uttley explained that traditional computers “can only consider one value at once,” a zero or a one. So for incredibly complex problems, you have to “do the next best thing … simulate and guess” to come up with workable solutions.
Using quantum computers, multiple values can be considered simultaneously, making “problems that are intractable today” seem less daunting, Uttley said. Honeywell and its partners have tasked “hybrid mathematician physicists” to create the computer, which employs a small ion trap — a device that has a tiny field of charged ions encased by electromagnetic fields, created by literally shooting the ions with a laser. These encased ions are called quantum bits, or qubits. It sounds very James Bond, we think.
Honeywell uses the quantum volume system to measure how well the computer operates, an IBM measure that takes into account how many qubits there are, as well as metrics such as error rate and time spent on calculations. IBM’s computer has a quantum volume of 16 as does Honeywell’s; Uttley said the company is hoping to soon reach 64.
Honeywell’s computer is also different from those available because others employ superconducting qubits, which are produced by cooled down electrical circuits. These are easier to produce than Honeywell’s method, but are less accurate and cannot be stopped mid-calculation. Uttley told New Scientist: “We can stop the calculation, take one qubit, ask ‘what are you right now, are you a 1 or a 0?’ and change the rest of the calculation based on that answer. It’s like putting an ‘if’ statement in an algorithm, and it’s something that’s unique to us.”
So how does a company like Honeywell go from making your home the ideal temperature to potentially solving issues like fraud detection or using artificial intelligence for partners like JP Morgan Chase? As Uttley said, the company was always in the “complex system integrator” business, and quantum computing takes that to a whole new level.
The issue now is finding applications for the quantum computer, available to customers within the next three months. It can be a challenge not just to build one but to also program it effectively. “We have had a number of partners kick the tires on our system,” said Uttley. Announcements about other uses are likely to emerge over the next few months.