Why not a drone? Some reasons for and against manning the next strike aircraft.
As I noted last Friday, Northrop Grumman has been awarded development of the US Air Force’s hoped-for Long-Range Strike Bomber (LRS-B), and the political-industrial campaign to kill the project is on. Unsuccessful teamed bidders Boeing and Lockheed Martin may each see a commercial interest in seeking a premature death for the effort, but with different hoped-for outcomes. Lockheed has its F-35 program to defend, and so may be happy to see the entire thing just go away. Boeing will be building tankers to refuel whatever aircraft the USAF buys (unless and until it loses a KC-Y competition to Airbus), but seeks no easy exit from the combat jet business. Rather, by challenging the fundamental requirement for a big manned bomber, Boeing could seek a do-over in a drone competition. The company may have a military argument too. Over months of researching the need for the LRS-B, I have heard arguments from plenty of industrialists, technologists, and strategists that the next bomber should be ab initio unmanned. The next bomber could be a drone, but there are reasons, good and bad, that it probably won’t be.
As Jonny Bardal of the Norwegian Defense Research Establishment argued in September at the 2015 Common Defense conference, autonomy has fundamentally become “a driver for cost-effectiveness,” and thus almost the school solution for problems in surveillance and targeting. For the US Navy, there’s thus a stark appeal to DARPA’s Sea Hunter, the robo-frigate known as the “anti-submarine continuous trail unmanned vehicle” (ACTUV). Trailing a diesel submarine with an unmanned 140-ton catamaran costs a fraction of sending a crew of 300 aboard a 9,000-ton destroyer. Similarly, when the Navy decided it needed a new aerial refueler to ease the buddy-tanking burden on its fighter-bombers, UCLASS was out, and CBARS was in. The Unmanned Carrier-Launched Aerial Surveillance and Strike drone was deemed a little too expensive and technologically ambitious, so the Navy is instead aiming to buy the RAQ-25 Carrier-Based Aerial Refueling System—a shipboard drone tanker that can double as a surveillance jet. The RFP will be released later this year, and will be limited to the four companies that had participated in UCLASS: Boeing, General Atomics, Lockheed Martin and Northrop Grumman.
Take Northrop Grumman’s own prototype UCLASS, arguably the most fully-baked amongst the four entries. At the cruising speed of an F-18 or F-35, a weaponized X-47B could have a combat radius of 1,200 to 1,500 nautical miles. That’s rather less than the widely presumed target range of 2,000 nautical miles for the LRS-B, but the gap may not matter too much. In a war with China (again, really the only reason for an LRS-B), this range would allow for staging from much of Japan, as even Tokyo is only 950 nautical miles from Shanghai. That does not get the plane the 1671 miles from Guam, but a single aerial refueling would. As Greg Knepper and Peter W. Singer pointed out last year, flying tankers may not be safe from long-range Chinese fighters, but a somewhat bigger drone could obviate that problem too.
If the target cost were as much as $100 million—and that’s a seriously expensive drone—the USAF could buy a squadron of six for the forecast price of each LRS-B. The threat of Chinese missile bombardment figures largely here. More, smaller planes could be more widely dispersed, more readily safeguarded in hardened bunkers, and more easily camouflaged on the ground. As inevitable combat losses occur—expect them to find some way of killing our planes—smaller, simpler aircraft could be more readily replaced. Downed drones also require no search-and-rescue, and can even be engineered to self-destruct to preserve their secrets. So with all this potential for robotic solutions, why does the LRS-B need an aircrew?
Actually, there are several reasons—organizational, sociological, and a few technological—for why this project is not now a drone. The first is lamentable, the second may be questionable, but the third set requires real answers.
The first is that sitting in a cockpit sits at the core of the culture of the Air Force. Narrowly, the brass may not have sought a fight with every bomber pilot in the service. Even the admiralty has been experiencing serious resistance from naval aviators about introducing a drone onto their carrier decks, and flying is hardly the only thing the Navy does. I might exclude the USAF’s missiles and space people, as they have never really sat at the cool kids’ table. Like the drone guys, they always look a little silly in flight suits and scarves. But as an Air National Guard general put it to me a few years ago, without big bombers and big missiles, the Air Force can start to feel a lot like an Army Air Corps. Or just the RAF. Either way, unmanned just doesn’t seem as manly.
Back to that “strategic” thing: the USAF does want the LRS-B to carry nuclear weapons. Leaving aside for the moment the question of whether a nuclear-armed, penetrating bomber is a good idea, that’s not the sort of mission entrusted to robots—except when it is. In November, Rachel Rizzo and I noted that an early opponent of this whole plan was “bomber-hating” General James Cartwright USMC. When Vice Chairman of the Joint Chiefs back in 2011, having previously served as Commander of Strategic Command—commanding all the nuclear weapons—Hoss influenced Defense Secretary Robert Gates to terminate the preceding Next Generation Bomber project. Back then, he told Time magazine that “nobody has showed me anything that’s required a person in that airplane—nobody.” Cartwright has even remarked that ballistic missiles are not manned either, and yet are trusted to deliver nuclear weapons. Yet for all the apparent reasonableness of that argument, there’s still something slightly off for most folks about drones with nukes.
Perhaps that’s because we fear them coming back at us. At a conference some months ago, I heard a war college professor assail the F-35 as “ultra-cyber-vulnerable.” That seems bombastic, but if the claim doesn’t seem preposterous, it’s because the worry derives from the seeming unknowability of anything buried deep in computer code. The Iranians may have already messed with the presumably secure midcourse navigation of another American drone. More measurably, there are communications problems with running massive drone wars, and Russian demonstrations of electronic warfare advances in Ukraine have been pretty impressive. The bandwidth challenge may eventually get resolved, but for now, the USAF thinks that its JSTARS replacement also shouldn’t be unmanned.
Similarly, if one wants the LRS-B to have the near-term potential for air-to-air weapons, then a manned flight deck may be not just a nice-to-have, but a necessity. Turning towards the ground, as Rachel and I argued late last year, mobile targets remain challenging for all aircraft, manned or unmanned. Perhaps the LRS-B’s potential for hunting down mobile Chinese missiles is debatable, but the worst case is definable. Loitering drones could attack moving targets autonomously, but whether they could discriminate amongst tanks, decoys, cargo trucks, and school buses is an important question. This gets back to sociological aversion to flying killer robots—an argument which will expand as the debate on the ethicality of autonomous robots continues. In defending its program, Northrop Grumman will have its own good arguments, because you don’t have to be August Cole to tell a Congressman the scary version of that story.
James Hasík is a senior fellow at the Brent Scowcroft Center on International Security.