JOINT BASE ANDREWS, Md.—More than 1,600 personnel, 60 companies, over a dozen aircraft, and dozens of radars and sensors came together for a massive Air Force-led event that utilized everything from boots on the ground to satellites in space to test how the service expects to fight a war in the future. The event culminated when a cruise missile was downed by a “smart” bullet.
The Air Force’s second “Advanced Battle Management System onramp,” long delayed because of COVID-19, took place on Sept. 3, largely within four national training ranges, as well as a part operations center, part fusion cell at Joint Base Andrews, Md. The new cloud-based, artificial intelligence-fueled command center brought together new and legacy ways of fighting wars into the service’s push for joint all-domain command and control.
“The focus is showing we really are building an internet for the military that feels like the internet that we use when we go home, except the things that we’re connecting are very different than the refrigerators, televisions, smartphones. They are warfighting systems and the operational need to move data quickly in a way operators understand really came out,” said Will Roper, the Air Force’s assistant secretary for acquisition.
The exercise played out with old and new sensors tracking a BQM-167 cruise missile, fired by USAF bombers posing as enemy aircraft, over White Sands Missile Range, N.M. The sensors fed into the real-time nerve center, where Airmen in the service’s new 13 Oscar Air Force Specialty Code gave the order for an Army Paladin Howitzer to fire the new Strategic Capabilities Office-developed Hyper Velocity Gun Weapon System, downing a cruise missile for the first time.
“Just for the record, tank shooting down cruise missiles. That’s just awesome,” Roper said. “That’s video games, sci-fi awesome. You’re not supposed to be able to shoot down a cruise missile with a tank. But, yes, you can, if the bullet is smart enough, and the bullet we use for that system is exceptionally smart.”
Playing War
The exercise focused on a hypothetical attack on the homeland—a culmination of “three months” of pretend increasing tensions that actually played out over the morning of Sept. 3. A “peer adversary” started a problem with the U.S.—since the scenario involved a real-world nation and is based on known capabilities, the event’s planners wouldn’t provide exact details. In response to this problem, the U.S. military begins to posture forces, causing the other nation to ratchet up its activity and firing the first shot, through a cyber attack.
The scenario then moved into space where real-world U.S. military assets used lasers to target satellites. In the Gulf of Mexico, adversary ships started an attack and drones targeted a U.S. Strategic Command convoy on the ground at Nellis Air Force Base, Nev. U.S. military assets took advantage of various sensors at multiple locations. For example, the personnel at Nellis had trackers live cameras, and even a robotic dog. An Arleigh Burke-class destroyer underway in the Gulf of Mexico, and USAF aircraft such as an E-3, E-8, F-15, F-16, and MQ-9, among others, were in the air across the country, with some aircraft scrambling on alert in Florida and flying at ranges in Nellis and New Mexico. At White Sands, legacy radars, infrared imagers, and sensor towers were combined with new systems to watch for threats.
The main live-fire scenario took place at White Sands, where bombers launched six BQM-167 targeting drones to simulate a cruise missile threat. Multiple systems targeted the BQM-167s, including the HVP round from the Paladin and a U.S. Navy deck gun, along with an AIM-9X Sidewinder missile fired from an F-16, an MQ-9, and a ground launcher. The outcome of the other launches is not public, Roper said, but the HVP’s success is and was a success story for a relatively unknown capability started at the Strategic Capabilities Office in 2013.
“It took a lot of selling to the Pentagon and to Congress, that hypervelocity guns could take on a variety of threats at a very low price point with a very high magazine to be a disruptive defense mechanism,” Roper said. “We were able to put it at center stage today, and it was successful.”
Andrews served as the focal point, where the entire operation played was projected on a large screen as well as tablets passed to VIPs. Cloud-based software programs combined all of that sensor and tracking data into one accessible screen, reminiscent of a real-time strategy-based video game. The software combined the live information with current readiness data at U.S. Air Force bases across the country, enabled by a real 5G tower providing massive amounts of bandwidth.
For example, clicking on Joint Base Elmendorf-Richardson, Alaska, showed how many F-22s were currently on alert, how many were fully mission capable, along with fuel and weapons available. Moving the screen to the adversary country showed their bases, with space-based and other intelligence data showing the most recent number of aircraft they have available and the recent pattern of life. The data ranged from what’s going on in space, down to individual people in the convoy in Nevada.
AI-focused applications, including the Pentagon’s Project Maven, combined the data from what’s currently happening along with readiness data to suggest courses of action for a commander to take.
This event—the second in the Air Force’s series of ABMS onramps—was planned by U.S. Northern Command and U.S. Space Command to reflect a mission set they need to be ready for. The goal of ABMS is to bring together all sensor data and provide a real-time view of what’s happening, and what decisions are possible.
“[The commands] are designed to match up against a world that didn’t have a peer threat, that was able to hold the homeland at risk,” a NORTHCOM official involved in planning the event said. “And that’s really what we face today. And so we have a contested environment inside of our area of responsibility for NORAD and NORTHCOM. … Space is rapidly becoming a contested environment that needs a combatant command with the capabilities to deal with those threats.”
How It Is Today
Currently, this is all done by phone and PowerPoint slides. At the North American Aerospace Defense Command headquarters, it takes a 12-minute game of telephone to get together relevant information if a possible threat develops. Someone watching radar sees a threat approaching North America, and picks up the phone to report there’s something out there and its speed, altitude, and if whether they are squawking a code. That information goes into a chat service, and other personnel report in what they are seeing and what assets are available to respond. “None of them are looking at the same picture, none of them have a visual display of what’s going on,” the official said.
All this information is then presented to an O-6 by phone, who then has to make the decision about whether the homeland is actually at risk.
“There’s no live collaboration in any common environment,” the official said. “The data is stove piped. The sensor data doesn’t talk to the Blue Force readiness data, the number of aircraft on alert, how much fuel is at certain locations. The only time all of that data comes together is in the mind of the O-6.”
To get some of this readiness data is a long process. NORTHCOM would need to reach out to U.S. Transportation Command if a scenario is developing to ask about the amount of tankers available. During a “dry run” of the exercise on Sept. 1, NORTHCOM and TRANSCOM were working together in the same program to speed up this process.
“Isn’t this how DOD works already? No. We have PowerPoint slides,” a separate NORTHCOM official said. “That’s literally how these kinds of decisions would be made. Right now, PowerPoint slides will be generated for commanders and for operators. … This process takes days to go through. What we showed … was the first time that combatant commands were in the same data cloud architectures and made decisions about posturing forces, and we did it in seconds.”
For another real-world example of the convoluted process that is currently in place, picture the Air Force’s Combined Air Operations Center in the Middle East. There’s a massive room with dozens of screens up on a large wall, and dozens of desks full of analysts watching different domains. All of the screens have different sets of information, with no way to view everything together.
“It takes 100 people to make sense of all those different feeds. And because it takes 100 people, you have to play 100 different telephone games to be able to go: OK, I’ve got a threat,” the NORTHCOM planner said. “If you’ve ever worked in a large organization, or [with] a huge staff, decisions don’t happen quickly because it takes a long time to get that information together.”
At Andrews, the new system came together in a converted conference room. The 13Os, or multi-domain warfare officers, lined up with their terminals running the ABMS systems, and behind them were lines of desks with experts from some of the companies participating.
“This capability did not exist six weeks ago,” the official said. “And as we are putting in more and more courses of action based on what the adversaries are doing, it is starting to learn this is how the humans like to do these things. And, over time, we create kind of a collaboration between man and machine on the best courses of action that can be created to be able to ensure our ability to go out and deter them.”
As the scenario developed, and an issue arose with the software, a company’s coder could jump in to find the issue and quickly push out a patch to fix it. During the Sept. 1 practice run, a program developed latency due to the large amount of information coming in. While the first ABMS onramp was done in an unclassified environment, the second scenario was secret and had to run through SIPRNet. The large amount of data was moving slowly, causing a delay of up to five minutes on some of the feeds. After some code work, the issue was patched.
To be able to participate in the ABMS onramps, and possibly move forward with actual programs of record, companies needed to play by the Air Force’s rules and program for one common architecture. The Air Force was able to bring in data from old systems, such as Link 16 communications, with new participants such as AT&T and SpaceX developing their systems to speak the same language.
“We are building a common architecture and we want to plug in both legacy sensors, legacy C2 systems, and be able to plug in future procedures,” the official said.
Going Forward
Because of the extended COVID-19-related delay, the next ABMS onramp is scheduled for about two weeks from now in Indo-Pacific Command. The follow-on, scheduled for early 2021, will take place in Europe and officials expect “Five Eye” allies to participate.
Some of the capabilities demonstrated in the second onramp have been in development for awhile and have shown they are ready for operations. CloudOne, for example, has been in the works for two years and has already proven itself, Roper said. Such capabilities could be brought on with indefinite-delivery, indefinite-quantity contracts moving forward.
“Certainly, the ability to utilize a cloud to share information is there and we need to move quickly and rapidly down that path,” NORTHCOM boss Gen. Glen D. VanHerck said. “I was very encouraged by the status, if you will, of those systems and the capability. I think they can be brought online within a year or less. It’s a matter of getting through the challenges with the department to field them.”
Some of the other components, like the artificial intelligence-focused tools, have more work to do. Data links in remote areas in the field went down. These failures are good, Roper said, because it outlines areas to improve. If nothing failed in the onramp, that means they weren’t trying hard enough.
For hardware, some of the sensors performed well and could be moved into NORTHCOM’s repertoire quickly, VanHerck said. The sensors would be fielded alongside current programs of record to supplement the capability at first to build trust and confidence in them, he said.
The onramp showed the Air Force can “project forward” its detection capabilities, with sensors currently available from commercial vendors, and combine that information in a way that enables the military to quickly decide if there is a threat and how it can be killed, VanHerck said.
Having those advanced sensors fused into a common operating picture as developed in the onramp will translate to more effective deterrence and alert missions within NORTHCOM, as some of the ideas and capabilities can come to fruition, VanHerck said.
“In the competition phase, you’re able to play checkmate because you have the information advantage and deterring the adversary is the end result. When you get into crisis, the same thing,” he said. “And so if you have an information advantage over the adversary, you’re able to quickly posture yourself and make decisions at the strategic level to the tactical and operational level. For me, as an operational commander moving forces, that would then deter any potential adversary in a more timely manner. So that was crucial.”
Editor’s note: This story was updated at 11:39 a.m. Sept. 5 to correct the name of the AIM-9X Sidewinder used in the test.