AI Adopts A Bigger Role In Space Operations

AI is destined to play a critical role in space if the opening of a new testbed center for space operations by aerospace giant Lockheed Martin is any indication. Called the Operations Center of the Future, the facility will allow satellite operators to manage multiple missions simultaneously using AI and secure, web-based cloud infrastructure. A key aspect to the technology will allow remote access for operators so they are available 24/7 in needed. The new operations center is located on the company’s campus near Denver, Colorado.

“The Operations Center of the Future next-generation AI, automation and cloud capabilities enable operators to remain closer to the mission than ever before, regardless of their physical location,” says Maria Demaree, vice president and general manager of Lockheed Martin’s National Security Space business. “Remote operators can receive timely mission alerts about satellite operations and then securely log-in to make smart, fast decisions from virtually anywhere.”

A key enabler is the company’s Compass Mission Planning and Horizon Command and Control software which already has been used to fly more than 50 spacecraft on a variety of government, research and commercial missions. This software allows for the management of a single satellite or a constellation of satellites by a single operator. Proof of concept came in April during the in-space mission of the LM LINUSS demonstrator, essentially a toaster-sized CubeSat that can rendezvous with a satellite already in orbit to provide servicing or upgrades. AI was used to coordinate flightpaths. The CubeSat also was able to maintain a secure, cloud-based link for mission telemetry, tracking and control. The new operations center is expected to be used for command and control of upcoming Pony Express 2, TacSat and LM-400 missions.


“Agile development, cloud-based operations and smallsat platforms came together at speed and in orbit where the real test of technology occurs,” noted Johnathon Caldwell, vice president and general manager, military space for Lockheed Martin.

The new operations center allows operators to use automation and AI/machine learning capabilities to manage satellite constellations of any size. These space constellations are increasing in size. The high-profile Starlink constellation, for example, is comprised of over 5,000 small satellites in low earth orbit (LEO).  Likewise, U.S. Space Command, also based in Colorado, is looking to use AI to keep track of orbiting objects that number over 46,000 in total. AI also is becoming more important in earth observation missions like the Space Edge Computer launched by Spiral Blue of Australia because of its ability to perform real-time analysis of earth observation data.

The use of AI, however, is likely to see a reduction in staff as Lockheed Martin notes the center requires only minimal staffing as operators all use the same baseline software and have the ability to work remotely.

AI, meanwhile, may soon be venturing further into space than LEO. Hailed as the “holy grail of astrobiology,” the new AI algorithm developed by university researchers can reportedly determine if material collected on another planet is biological or not to within 90 percent accuracy, sometimes even if the sample is hundreds of millions of years old. Instead of looking for specific things that might indicate life, the AI looks at the differences between samples. The AI algorithm works off the premise that the chemical processes in the formation of biomolecules fundamentally differ from those from abiotic sources. The AI was trained using 134 biotic and abiotic samples until it was able to differentiate between living things like shells, teeth, bones and hair and fossilized fragments made of things like oil, coal and amber. Mars would likely be the first off-world destination for its use, although it may also be used to resolve questions surrounding samples already collected.

“It opens the way to using smart sensors on robotic spacecraft, landers and rovers to search for signs of life before the samples return to Earth,” said lead researcher Professor Robert Hazen of the Carnegie Institution Geophysical Laboratory. If alien life were to be found, AI could determine whether it derived from a common source or was of different origin.