WASHINGTON – A project birthed in the Pentagon’s Strategic Capabilities Office is getting some serious buy-in from the U.S. Navy, and could yield the world’s first large-scale armed unmanned warship.

The Navy raised eyebrows in its budget rollout Tuesday when it requested $400 million for two large unmanned surface vessels to be purchased in 2020, with 10 total to be purchased across the five-year projection known as the future year defense program. But it was not immediately clear what exactly the Navy was buying two of, since no program of record exists for a large unmanned surface vessel (or LUSV).

Navy officials now say the request is an outgrowth of SCO’s Project Overlord, which first surfaced in 2017 with a draft solicitation outlining a program that would take existing autonomy technologies and integrate them into large and medium unmanned surface vessels with some heady ambitions: an autonomous ship capable of carrying up to 40 tons of payloads, and operate in up to sea-state five independently for 90 days without a crew for maintenance, while following all rules of navigation and obstacle avoidance.

The elevation of SCO’s Overlord program from science project to fast-tracked acquisition reflects both the Navy’s growing confidence in its ability to make the technology work and the urgency it feels to field technologies that can combat a growing threat from anti-access area denial technologies designed to keep the Navy’s powerful strike arm far from the shores of potential adversaries.

It’s an effort the Navy is ready to put some significant investment into. In total, the service has programmed $2.7 billion across the FYDP. And on Wednesday evening, the Navy dropped a request for information from industry seeking to “determine if sources exist that are capable of satisfying the Navy’s anticipated program requirement for Large Unmanned Surface Vessels (LUSV).”

The Navy’s top officer Adm. John Richardson said he wants the force to move aggressively on getting the LUSV to the fleet, and compared it to the unmanned aerial refueling drone, the MQ-25, one of his top priorities as chief of naval operations.

“I would liken this to the surface vessel version of where we picked up on MQ-25,” Richardson told a roundtable of reporters on Wednesday. “We are moving very aggressively to get something on deck in unmanned aviation, and we were able to do that very quickly by taking advantage of what we’d learned in that field to date, bringing industry in early, so we’re going to be using a lot of those practices.

“This seems like kind of the next natural step. We want to move it out of the skunkworks phase and into the operational phase as soon as possible.”

Richardson pointed to recent strides the service has made in the unmanned surface vehicle realm with the Sea Hunter program, which recently transferred from the Defense Advanced Research Projects Agency (DARPA) to the Office of Naval Research, as evidence of the Navy’s progress toward fielding unmanned warships. The LUSV program, however, is a separate effort.

The Navy announced in February that Sea Hunter navigated autonomously to Hawaii and back.

Overlord

The Overlord program from which the Navy hopes to derive its new large unmanned robot warship is the child of the Pentagon’s Strategic Capabilities Office, a workshop under the auspices of the Office of the Secretary of Defense that seeks to add funky new capabilities to existing gear and weapons.

The office’s big headline-grabbers to date have been its work on drone swarms and converting the Navy’s SM-6 anti-air missile into a very long range ship killer.

But its work on autonomous surface vehicles has been less front-and-center.

According the 2017 draft plan, Overlord is seeking a ship that can do virtually everything a larger manned vessel can do – obey the international rules of the road for navigation, plan a route for a mission, communicate with other ships (manned or otherwise) in a task force – and do it with very little interaction with sailors once it gets underway.

“The Overlord program will develop core autonomy, communications, and C2 components and field prototype USVs capable of being seamlessly operable with the fleet,” the draft says. "The Overlord program will have built in redundancy in all critical hardware and software systems. The program will involve integration and test of payloads for [electronic warfare], [anit-surface warfare], and [strike warfare].”

The program, in keeping with SCO’s modus operandi, places an emphasis on using and adapting existing vehicles and technologies, the draft says, and “will take advantage of commercial technologies, integrate existing vehicle designs, and mature existing autonomy capabilities to accomplish its goals.”

The draft lays out a two-phase plan that starts with a 12-month phase I, which asks industry to demonstrate a vehicle that could meet the requirements laid out in the draft, including a range of 4,500 nautical miles, “capable of operating in at least Sea State 5, with at least 80,000 lbs. of payload capacity and 75 kW of 450V, 60 Hz, three-phase AC power reserved for payloads."

The second phase of the program is a down-select to the most competitive entries for further testing and integration, shifting from the unclassified phase I to a “Secret”-classified phase II.

Naval Sea Systems Command declined to respond to questions about the progress of Project Overlord, citing operational security, but acknowledged the LUSVs being acquired in the 2020 budget are an outgrowth of Overlord.

The new LUSV will be critical to operating in a distributed way in the future, NAVSEA spokesman Alan Baribeau said.

“LUSVs will provide relatively low-cost, high-endurance, reconfigurable ships able to accommodate various payloads for unmanned missions to augment the Navy’s manned surface force,” he said in a statement.

The request for information released Wednesday that seeks more inputs on vehicles from industry is continuing of the effort started with Overlord, said Danny Hernandez, a spokesman for the Navy’s Research, Development and Acquisition office.

“Overlord Vessels were experimental efforts," Hernandez said. “The RFI is a separate effort from Overlord that will launch the LUSV prototyping project."

Moving fast

In the Navy’s fiscal year 2020 budget rollout, the Navy’s budget director told reporters that the large surface combatant, part of what the Navy is terming its “ghost fleet,” would be about a third the size of the FFG(X), which could displace upwards of 6,000 tons.

“They are 200 to 300-foot, 2,000 ton [hulls],” said Rear Adm. Randy Crites. “I’m not sure what the final hull form will be, that’s what we’re using today in terms of what the ‘ghost fleet’ buy is. But I don’t think we know yet exactly what the hull form will be.”

The Navy is adamant that, even without a fully fleshed out requirement, it needs to move fast to get these capabilities into the fleet.

“We've got to get through the details of concept of operations, command control – how is it going to work in a distributed environment,” Crites said. “But we need these test articles and we need to bring these things on quickly so that we can actually see how this is going to work.”

The path the Navy is treading could upend how the fleet has fought since the Cold War. The CNO’s surface warfare director previewed what the surface fleet has in mind in a December interview with Defense News.

Boxall laid out a vision for pushing as much capability to smaller, distributed and unmanned ships as possible, and using the larger ships as command and control for the ghost fleet.

“Today, I have a requirement for 104 large surface combatants in the force structure assessment; I have 52 small surface combatants,” said Surface Warfare Director Rear Adm. Ronald Boxall. “That’s a little upside down. Should I push out here and have more small platforms? I think the future fleet architecture study has intimated ‘yes,’ and our war gaming shows there is value in that.”

The paradigm shift is moving the fleet away from platforms like the Arleigh Burke-class destroyers — enormous, tightly packed ships bristling with capabilities, weapons and sensors, but enormously expensive to build, maintain and upgrade.

“It’s a shift in mindset that says, instead of putting as much stuff on the ship for as much money as I have, you start thinking in a different way,” Boxall said in a December interview. “You start saying: ‘How small can my platform be to get everything I need to be on it?’

“We want everything to be only as big as it needs to be. You make it smaller and more distributable, given all dollars being about equal. And when I look at the force, I think: ‘Where can we use unmanned so that I can push it to a smaller platform?’”

“Distributed” is the key word in a new concept the Navy is developing to counter China’s increasing grip on the South and East China seas. Though public details of what the Navy is calling “distributed maritime operations” are scant, it seems to stem from an idea that was developed inside the surface Navy, known as “distributed lethality.” The idea was to put the Navy’s surface combatants back on the offensive, spreading out to stretch Chinese intelligence and surveillance assets and leave openings for offensive strikes.

NAVSEA said in its statement that the LUSV would be a key component of distributed maritime operations.

“LUSVs will provide relatively low-cost, high-endurance, reconfigurable ships able to accommodate various payloads for unmanned missions to augment the Navy’s manned surface force," Baribeau said. “LUSV is a key enabler of the Navy’s Distributed Maritime Operations (DMO) concept, which includes being able to forward deploy (alone or in teams/groups), team with individual manned combatants, or augment battle groups with an LUSV-distributed Anti-Surface Warfare (ASuW) and Strike capacity.”

Secure, Fast and with Enough Bandwidth?

That kind of integration with the fleet – where lethal capabilities heretofore reserved for manned warships are outsourced to robot ships controlled by humans on larger crewed vessels – is integral to the Project Overlord vision as laid out in the draft proposal.

But it’s also something that is immensely hard to pull off, especially in a contested environment where communications will be both targeted and exploited by adversaries looking for targets, experts say.

“You could communicate via satellite, but then that’s vulnerable to the same anti-access, area-denial (A2AD) disruptions that you would normally have to worry about,” said Michael Horowitz, an expert in unmanned systems and a professor at the University of Pennsylvania.

“You could have localized sensor nets. Then you need a way to link that data back to the human-piloted system. You need some way to get the information back to humans and that’s going to be a key vulnerability.”

There is also a problem with the sheer volume of data you would need to transmit over nets, presumably relayed through either communications relay drones or other ships (nodes) in the network, Horowitz said.

“The weak point is whether you can extend your sensors – your robotic connectivity,” Horowitz continued. “If you think about it like a chain that, say, goes all the way back to a carrier that’s really far away, or a cruiser. … Remotely piloted systems, that requires a lot of bandwidth. The more nodes you have, the more you increase the risk that an adversary can disrupt you by knocking one of them out.

“So, you deploy it as a net so that any one node isn’t critical, but then you need to deploy a lot of them.”

And scale is a problem because none of this technology is cheap, Horowitz said.

“It requires ever-greater miniaturization, in terms of munitions, power, etc.,” he said, “depending on how small you want these platforms to be. If what you’re talking about is Sea Hunter size, well, the question is could you scale Sea Hunter today? I think the production costs would likely be pretty high.

“I think the key will be driving down the costs of producing some of these platforms.”

Still, the value of pursuing the vision has significant potential, he added.

“I think it makes sense to invest in this kind of capability,” Horowitz said. “Will it pan out in a way that is secure and fast and with big enough band width? These are the challenges.”