APOGEE STAFF
U.S. military leaders are counting on artificial intelligence (AI) to play a key role in managing and tracking the thousands of satellites in space while also helping to tackle complex threats that include GPS jamming, counterspace operations and space junk, to name a few.
Government agencies and private industry are poised to launch unprecedented numbers of spacecraft in coming years, with the Space Development Agency (SDA) alone anticipating hundreds of new satellites as part of its Proliferated Warfighter Space Architecture (PWSA) to support data transport, communications, end-to-end missile warning and tracking, and battlefield awareness and targeting.
As of June 2024, the European Space Agency estimated there are more than 10,000 active satellites in space, including about 6,150 that function as a part of Elon Musk’s Starlink constellation. But that’s only a shadow of what is to come as industry and government prepare to step up their launch schedules: Industry analyst Chris Quilty told Yahoo! Finance in December 2023 that he estimates another 20,000 satellites in space by the decade’s end. Musk’s ability to lower launch costs has made satellites more affordable and attractive to companies, spurring growth. “It’s lowering those launch costs that have suddenly enabled people to look at business models, whether it’s manufacturing in space or building a data center on the moon, and suddenly the economics work,” said Quilty, co-CEO and president of Quilty Space, based in St. Petersburg, Florida.
With the proliferation of spacecraft — and the ever-persistent threat to those networks by Iran, the People’s Republic of China (PRC), Russia, and others — government officials have their hands full. Experts, including Quilty, say the congestion will only worsen as more satellites enter orbit.
The task of observing the world’s spacecraft, including adversarial ones, falls to the U.S. Space Force. To optimize the role Space Force Guardians play in monitoring that traffic, military leaders are exploring the potential of AI and its subset machine learning (ML). Not only can AI categorize the thousands of satellites orbiting the planet in real time, but it can plot each by position, even as they hit speeds up to 29,000 kilometers per hour.
Also important, officials say, AI can help Guardians focus more attention on where it’s needed most: potential threats to U.S. satellites, including by the PRC and Russia. As National Reconnaissance Office (NRO) Director Chris Scolese noted in his address to the 2023 Space Symposium in Colorado Springs, Colorado: “Automation and machine learning can focus on the ‘what,’ ‘where’ and ‘when’ of intelligence questions and allow humans to focus on what they do best, answering the ‘why.’”
Adversaries determined to disrupt U.S. and allied satellite networks aren’t the only threats Guardians face. They also monitor space junk — fields of derelict satellites, rocket boosters and other objects coursing through space — and warn operators when their satellites are at risk from it. AI could track junk that’s softball-sized or larger, which numbers an estimated 35,000 objects. U.S Space Command (USSPACECOM) says the number of objects it tracks, including debris, has almost doubled to more than 46,000 since 2019.
Officials haven’t said when AI will become a permanent fixture in military space architecture — satellites, networks and ground stations — or to what degree that could happen. However, its introduction is anticipated as more satellites are placed on orbit. Those numbers accelerate at a steady clip. The SDA has launches scheduled through 2030 to populate its vast PWSA constellation. The next group of satellites, Tranche 1, was set to deploy in late 2024, followed by successive monthly launches. Tranche 2 will launch beginning in fiscal 2027.
At the same time the SDA, an arm of the Space Force, builds its constellation, the NRO aims to quadruple the number of intelligence and reconnaissance spacecraft it operates within the next decade. Scolese said AI could become a boon to the agency as it expands its network. “When you’re operating one satellite, one person can sit there and think about it,” Scolese said in an April 2023 report in DefenseScoop. “When you start multiplying that to tens or hundreds, it becomes more important to go off and look at how I am optimizing that constellation.”
Tracking satellites and other objects isn’t important only to satellite operators, it’s vital to overall space domain awareness — a component of space defense. Space Force Gen. Stephen Whiting, commander of USSPACECOM, told the Senate Armed Services Committee in February 2024 that this is one area where AI could lend a hand. “USSPACECOM increasingly relies on a near-real time, comprehensive understanding of the congested and complex space operational environment. Space Domain Awareness data, operational C2 automation, and incorporation of artificial intelligence/machine learning capabilities will allow us to better understand and outpace the threat in our unique area of responsibility,”
he said.
Whiting also stressed the potential of AI in May 2023 when he spoke at a Mitchell Institute for Aerospace Studies event, asserting AI is already making inroads at USSPACECOM: “It is a practical application that we need to be driving across all of our operations and business practices,” he said during a segment of the institute’s Schriever Spacepower Series. “Inside our headquarters, Space Operations Command, we have a deputy commanding general for transformation organization, and that organization right now is working through, how do we implement a generative AI type of capability inside the headquarters to help us get after our business practices?
“We’ve also established a top 10 AI ML needs list across our operational missions. And working with partners like Space Systems Command and the Air Force Research Lab, we’re working to actually field capability, and we’ve seen some of that in our … space domain awareness system where we have a lot of data, and AI ML can help us parse through that data,” Whiting said. “We’ve also seen it in some predictive maintenance activity, predictive maneuver-type capability.”
AI could prove beneficial in other areas, including intelligence gathering, data transfer and enhanced connectivity to assist warfighters. In 2023, the Pentagon’s Chief Digital and Artificial Intelligence Office held a series of exercises called the Global Information Dominance Experiments (GIDE), which marked the first time the Department of Defense (DOD) had held a joint, globally integrated experiment enabled by data, analytics and AI.
The experiments, involving military and civilian personnel from all the service branches and multiple combatant commands, were carried out four times in 2023 with the goal to “inform Joint All-Domain Command and Control (JADC2) solutions related to joint data integration and the use of AI and machine learning technology,” a DOD statement read.
Of course, the U.S. isn’t alone in exploring AI. Recently retired Space Force Chief Technology and Innovation Officer Dr. Lisa Costa said during a November 2023 event at the Mitchell Institute that the PRC, identified as the military’s “pacing challenge,” is aggressively pursuing AI in space — and investing heavily in the technology.
A common problem that cuts across all of these areas, not only for China but for us, as well, is the amount of data that we have and the quality of data that we have. I fundamentally believe, though, that this is going to be a solved problem very quickly.”
~ Dr. Lisa Costa then chief technology and innovation officer, U.S. Space Force
The PRC’s aim, she said, is to “dominate in AI” to “probe enemy vulnerabilities and advance the concept of intelligentized war.” She added that the People’s Liberation Army envisions an orbital platform equipped to make decisions based on AI about who is and isn’t an adversary.
The PRC’s investment is substantial: $14.7 billion in the past year. “In two short years in FY26, that is expected to almost double to $26 billion. That’s an extremely large investment,” Costa said. “And not only are they applying that to AI research, they’re applying it to the operationalization of AI.
“A common problem that cuts across all of these areas, not only for China but for us, as well, is the amount of data that we have and the quality of data that we have,” Costa said. “I fundamentally believe, though, that this is going to be a solved problem very quickly, and why do I believe that? Machine learning and natural language processing is at a point now that computer-based tagging of information, large amounts of information in real time is possible and, in fact, computers are much better at tagging and marking up data than humans are. They’re much more consistent at least, and so I believe this is going to be a real game-changer in terms of being able to use AI in the operational space. The real key is, who will be able to implement reliable, secure and trustworthy AI?”