NASA’s pursuit of commercial hypersonic flight was just given an AI-powered boost

One-hour flights anywhere may be some way off yet, but artificial intelligence could play a massive part in speeding up the development of hypersonic airliners.

U.S.-based Argonne National Laboratory announced a partnership with NASA to boost hypersonic flight research and make vastly shorter travel times a reality with the help of AI-enhanced computer simulations, a press statement reveals.

Simulating hypersonic flight with AI

Hypersonic flight is achieved at a speed of Mach 5, or five times the speed of sound at sea level — sound travels differently at different altitudes and on different planets.

Argonne will bring its supercomputing capacity to the table to help NASA develop its hypersonic testing systems, including experimental aircraft such as its X-43A scramjet-powered aircraft, built as part of its Hyper-X program.

The company uses computer fluid dynamics (CDF) to model and predict how an aircraft will react to the forces around it at hypersonic speeds. In its statement, Argonne defines CDF as a scientific field devoted to "numerically expressing the behavior of fluids such as air and water."

Though there is a lot to discover about the effects of Mach 5 and above on aircraft, Argonne believes its supercomputing capacity will help to speed up the process. The lab presented a paper at an American Institute of Aeronautics and Astronautics (AIAA) Forum earlier this year, detailing how it's using artificial intelligence to improve computer simulations and streamline the development of state-of-the-art aircraft.

Dreaming of Mach 5 airliners

As Argonne points out, air friction at hypersonic speed is so strong that it can melt parts typically used on a traditional commercial aircraft. NASA and other organizations are testing scramjets, which take advantage of the fact that air is compressed at such high speeds, leading to greater fuel efficiency.

"Because the chemistry and turbulence interactions are so complex in these engines, scientists have needed to develop advanced combustion models and CFD codes to accurately and efficiently describe the combustion physics," said Sibendu Som, a co-author of Argonne's study.

Argonne said it successfully applied its own CDF to NASA's CDF, called VULCAN-CDF, and that it has shown promising results. The lab's scientists used machine learning techniques to reduce the large memory requirements and computational cost associated with simulating fuel combustion at hypersonic speeds.

All of this will help NASA better understand the effects of Mach 5 and above on aircraft, allowing it to continue to pursue the development of experimental hypersonic aircraft. We may have to wait a long while to see the benefits of hypersonic flight in any kind of commercial capacity, but Argonne believes its system will help greatly speed up NASA's pursuit of reaching commercially viable hypersonic speeds.