Inside the quantum challenge at Airbus: applying quantum computing to enterprise problems

An ongoing challenge from Airbus shows a potential path to take academic research into applied science to solve enterprise problems

Share

Airbus is seeking to pool together the best minds in quantum computing to push forward aviation design, optimisation and modelling with a competition that will run throughout the year.

One of the goals of the competition is to find areas where quantum computing can provide an advantage over more traditional high performance computing (HPC). The challenges are split into five problem statements, and these concern: aircraft climb optimisation, computational fluid dynamics, quantum neural networks for solving partial differential equations, wingbox design optimisation, and aircraft loading optimisation.

airbus quantum challenge istock 2019

Quantum computers are still in their infancy, but the aim is to elongate the length of time that quantum bits (qubits) can hold a stable state before starting to deteriorate. The longer this highly volatile state can be maintained, the more functions the quantum computer can effectively carry out, potentially offering vastly superior computation to that in even the most modern computers.

Dr Thierry Botter, head of Airbus's Blue Sky research lab, told Computerworld UK that the company had already been examining quantum in the context of aviation for a number of years - not least with an investment in QC Ware through its venture capital wing last year.

"We're being pragmatic about this, we're not asking the question: can it be done today? But rather can we see an indication that in some future there will be an advantage from quantum computers?" he asked. "How can we prepare ourselves today to be ready at that point in the future?"

This competition, then, extends that into an effort to engage with the quantum computing community "as widely as possible", he added.

"We laid down a number of problem statements all having to do with one of our core businesses, like flight physics, and essentially laid down the challenge to the community," said Botter, "to say: please help us out, please help us understand the promise, the potential of quantum computing, and how that new technology could be used to answer some of these problems."

At the time of our interview in March, Botter said that there were already 350 registered participants in the competition from 45 different countries, signalling a strong global interest.

"It makes us all the more excited for the end of the submission period," said Botter. "Right now, we're still in the early stages - individual teams are crafting their submissions, their ideas."

Potential participants can register their interests at the Airbus quantum challenge website here.

Take-off

Although the cruising segment of a flight lasts the longest, and is considered by airlines to be the most important related to fuel and time optimisation, the increasing prominence of short-haul flights, driven by greater competition from budget airlines, means that fuel optimisation during climb and descent are also incredibly critical to an airline's bottom line.

"Problem statement one has to do with the very first part of an aircraft to fly, it's going not from the runway to cruising altitude but from a starting point in-flight and still relatively near the take-off point of an aircraft," said Botter. "We want to map that leg of the climb, from the early stage of the flight all the way until the aircraft reaches cruising altitude - we'd like to optimise that segment of the flight."

Optimisation can be measured and calculated in various ways, but at its core reducing time or fuel consumption, or a combination of the two, is key.

"Can we optimise that early segment? That is the question," says Botter. "There are classical ways to tackle the problem, and we know them. That's not the point of this particular problem statement - it's to put ourselves in the future and ask the question: can quantum computers help us find an even further optimal solution to these problems? If so, how are we going to carry that calculation? What would we need to do in order to obtain an even better, even more optimal, solution to this challenge?"

Read next: Inside IBM Q System One: an archetype for the quantum computing frontier?

Participants submitting proposals are asked to put forward details of their processes and their methodology, as well as some metrics and indications of potential performance by running simulations at a smaller scale. The idea is that even with limited access to resources to actually run the problems, Airbus and the other judges will be able to gauge some level of the expected performance boosts.

Following the challenge, Airbus hopes to work with the winners directly, providing them with access to expensive quantum computing hardware so that they can test the problem statements out with real quantum technology.

Botter adds that although the quantum computers of the near future are not going to be as powerful as the error-corrected quantum machines of the future, they can "nonetheless potentially start to provide some added benefit".

"The question is how, and for what application? I think this is the big quest that we at Airbus and many others are engaged in, to find those key applications that will be able to draw already on the near-term potential use of these quantum computers, and further place us in a position of strength for when the future, error-corrected quantum computers come in."

Applying quantum

A major part of that will be taking a technology that has matured in a laboratory environment and been driven by academics, and translate it into the real world.

"Going beyond the bounds of aerospace there are other sectors also looking at how quantum computing can help them, from the finance sector to the energy sector, and many other transportation sectors," he said. "For all of them to get engaged and tap into the development that has already been done, that is emanating out of academic research labs, this is the big challenge we see globally outside of just Airbus."

Fundamentally Botter sees quantum computing as a potentially game changing technology, but only if academics and industry collaborate. 

"The engagement of as many actors, and as many end users as possible, is only going to improve and accelerate the development of quantum computers for all applications," he said. "It's one of those rare scenarios where there are only winners. Everybody has calculations that are rather challenging, and difficult, and traditionally very computationally hungry, and can benefit from having improved, better technologies - namely quantum computing - come in and help them with their calculations."

One area where quantum might be able to help (but that is outside the scope of this particular challenge) is in satellite vision optimisation. "Today we have a constellation of Earth-observing satellites, they each have their own characteristics in terms of imaging capability," he said. "They're on their own orbital track. Together they form an ensemble of potential imaging devices, whose type of imaging we want to optimise, and we want to map that as best as possible to a series of requests that are coming in from customers.

"To do this mapping in the most optimal way is again an optimisation problem. There are techniques today that can be applied, but we're already preparing the future and asking the question of how quantum computing can help - this work of trying to identify the different segments of potential applications of quantum computers to harness and improve existing HPC is a big, big push that we are after."

"Recommended For You"

Quantum computing revolution claim by academics Researchers use silicon to push quantum computing toward reality