• Rolls-Royce, Riverlane and Xanadu have secured more than £400,000 grant funding from Innovate UK.
  • The grant is aimed at accelerating the development of applications that will allow quantum computers to model the flow of air through jet engines.
  • The project, called CATALYST, will deliver a hybrid quantum-classical framework combination.

Rolls-Royce, Riverlane and Xanadu have secured more than £400,000 grant funding from Innovate UK to accelerate the development of applications that will allow quantum computers to model the flow of air through jet engines. An additional CAD $500,000 has been awarded from the National Research Council of Canada Industrial Research Assistance Program (NRC IRAP) as part of a growing relationship between the UK and Canada on quantum computing technology and expertise.

The project, called CATALYST, will deliver a hybrid quantum-classical framework combination, where computers of the type we use now are programmed to instruct quantum computers. This will give Rolls-Royce the means to rapidly evaluate and implement new quantum algorithms, accelerating the time to do this from several hours to just a few minutes. This will bring huge efficiencies to future product design processes and also contributes to the first of the UK Government’s recently announced National Quantum Strategy Missions.

UK-based quantum error correction company, Riverlane, and quantum computing company, Xanadu based in Ontario, Canada, are already existing partners with Rolls-Royce on other projects and share the long-held vision that only error-corrected quantum computers will deliver a broad and lasting quantum advantage.

Leigh Lapworth, Rolls-Royce Fellow in Computational Science, said: “This is the first quantum computing research and development collaboration to be led by a large industry partner, instead of smaller start-ups. Developing industrially robust software is a multi-year activity, even without the new thinking needed by quantum computers.”

“Our shared vision and approach will make us one of the first companies to benefit from fault tolerant computers. The techniques we develop in this project will be those that enable us to benefit from the UK’s quantum pathway of a million error-corrected quantum operations in 2028; a billion in 2032; and a trillion in 2035.”

CATALYST draws on the unique expertise of each partner: industrial applications (Rolls-Royce); quantum algorithms (Riverlane); and hybrid quantum-classical compilation (Xanadu).

Steve Brierley, CEO and Founder from Riverlane, said: “The CATALYST project brings together leading quantum computing companies and industry experts from the UK and Canada to help improve the quality of the quantum algorithms. By developing better quantum algorithms, we can reduce the number of quantum operations required to unlock world-changing applications, sooner. Such work across the quantum computing stack is vital to help us unlock millions and then trillions of reliable quantum operations, aligning with the UK Government’s targets to reach the so-called MegaQuOp threshold by 2028 and TeraQuOp threshold by 2035.”

Josh Izaac, Director of Product at Xanadu, said: “As quantum hardware continues to grow in both scale and capabilities, we need to re-think the quantum software technical stack to enable the design and execution of larger and more complex quantum algorithms.”

“Working alongside Rolls-Royce and Riverlane, world-class leaders in algorithm development and quantum error correction, will guide the design and development of Catalyst, our just-in-time quantum compilation framework. This will unlock the ability to explore bigger, more complex, and more dynamic quantum algorithms with PennyLane and our world-class simulators.”

Rolls-Royce’s largest research models solve equations with several billion variables and its ambition is to reach a trillion and beyond. It will be a decade or more before quantum computers achieve these scales, but experience shows that the path from here to there requires many problems to be solved, each one depending on the solution of its predecessor.

Many of the processing steps are, and always will be, on a classical computer. These can dominate the end-to-end run time to such an extent that, if unaddressed, they will prevent even an infinitely fast quantum computer from delivering a meaningful advantage. Accelerating the time taken to do this will enable users to investigate more demanding test cases and develop the tools and software needed to demonstrate industrial advantage.