In just over 18 months--if upcoming flight tests are successful--the Air Force will begin setting formal requirements for robotic attack airplanes. Highly stealthy and equipped with a variety of sensors, these Uninhabited Combat Air Vehicles of the year 2008 will operate well behind enemy lines, sniffing out hidden air defenses and swiftly destroying them. They will also strike targets ringed by the most lethal surface-to-air missile systems and likely carry out a variety of other missions, from surveillance and reconnaissance to jamming.

Eventually, they will also be the first "laser fighters."

Initially, these aircraft will supplement the manned strike fleet but could later replace what service leaders term a "significant" portion of it. And, far from clinging to a "white scarf" mentality that sees UCAVs as a threat to the livelihoods of pilots, service leaders have had to rein in their exuberance for the new class of aircraft, lest they get ahead of where the technology really is.

"We plan to pursue this program once the [advanced technology demonstrations] are over," said Gen. John P. Jumper, head of Air Combat Command. "I don't think there's any doubt about that. ... UCAVs will come, and we will work the concept of operations to include them."

Maj. Gen. (sel.) David A. Deptula, Air Force National Defense Review director, told Congress in March that UCAVs would be one of "four platforms [that] will define the stealthy Air Force of 2020," alongside the F-22 fighter, B-2 bomber, and Joint Strike Fighter.

Two recent events underscored the fact that UCAV technology is moving ahead rapidly.

First, the Air Force this spring demonstrated that the Predator reconnaissance drone could destroy a tank. It equipped the aircraft with a laser designator and a Hellfire missile typically carried by Army helicopters. The Predator fired shots in a series of tests from low level and then from 15,000 feet, Predator's normal operating altitude. While the service has no plans to buy Predators for such a mission, the experiment was deemed a good first step in working through the challenges of using robotic aircraft to conduct attacks. Additionally, the Air Force now knows it could press Predator into a strike role if a situation warranted it.

Down Under and Back

In the second watershed event, the Global Hawk unmanned reconnaissance aircraft executed a totally autonomous, 30-hour flight from California to Australia. After taking off, the Global Hawk flew a representative surveillance pattern en route to Australia, acquired the landing airfield, and set itself down. Weeks later, it repeated the achievement in reverse, on its return to the US. Both flights were completely hands-off by human beings. The Global Hawk success indicates how far air vehicle autonomy has come in only a few years.

Advisors to the new Bush Administration have taken note of the advances. A panel exploring "transformational" strategies and technologies for Defense Secretary Donald Rumsfeld recommended heavy investment in robotics in general and Unmanned Aerial Vehicles and UCAVs in particular as a way to drive down both risk and cost in prosecuting future air wars. Reportedly, big shifts in funding emphasis highlighting robotic airplanes will surface in the Fiscal 2003 budget, scheduled for public release next January.

Congress, too, has climbed aboard the bandwagon. Sen. John Warner (R-Va.), now the ranking minority member of the Senate Armed Services Committee, said last year that he believed the Pentagon should "aggressively develop and field unmanned combat systems in the air and on the ground" and worked to add $146 million to the budget to speed development of UCAVs. Warner went on to say he expected that, "within 10 years," fully one-third of the nation's deep strike aircraft could be robot airplanes.

In an Unmanned Aerial Vehicles roadmap released in April, the Pentagon said that through the 1990s it spent some $3 billion on UAV development, procurement, and operations and expects to spend $4 billion in this decade. It also said that, by 2010, it will have more than tripled the number of fielded UAVs, from about 90 in all services today to some 300.

The Air Force, in conjunction with the Defense Advanced Research Projects Agency, is sponsoring a UCAV program initially focused on the Suppression of Enemy Air Defenses role. The SEAD mission was selected for USAF's initial foray into modern robotic air war because the service deemed existing solutions inadequate and the mission an especially risky one for pilots.

"The surface-to-air missile threat [and] integrated air defense threat is getting increasingly more difficult," said Maj. Gen. (sel.) Daniel P. Leaf, head of operational requirements for USAF.

Soft Kill, Hard Kill

The Air Force's current SEAD airplane, the F-16CJ, is more "reactive" than the service would like. It tends merely to keep enemy radars from turning on, rather than destroying them-an action known as a "soft kill" in military parlance. USAF wants to get a "hard kill"-destruction-on enemy air defenses, especially mobile units, so they don't chronically reappear or lurk through a war, adding complication and risk to an air campaign.

Use of a UCAV in this role not only decreases risk but also increases effectiveness because the vehicle is smaller and less observable than a manned aircraft. "It's very attractive," said Leaf. "It's a natural fit for UCAV."

Boeing won a competition over Northrop Grumman and Lockheed Martin to build the Air Force/DARPA X-45 UCAV. The unique, Y-shaped airplane, designed for stealth and able to carry thousands of pounds of ordnance, is expected to fly this summer. Over the next year and a half, it will demonstrate basic flying qualities in an autonomous mode. In a follow-on set of tests, it will be judged for its operational qualities, such as its ability to find targets with its onboard sensor suite, to fly in conjunction with manned aircraft, and demonstrate high operational reliability. It will also demonstrate attacks on ground targets with several types of weapons.

When that's all done-or maybe sooner, if the tests go well-the Air Force will begin planning a full-up UCAV that could enter development as soon as 2003, with an initial capability possibly as soon as 2005, but no later than 2008.

"There's a lot of interest inside ... and outside the Air Force in unmanned vehicle technology," Leaf noted. "It could always be accelerated."

Leaf was reluctant to say when formal requirements for an operational UCAV will be set, since the X-45 must first prove itself in flight test and the Pentagon must complete its ongoing review of programs and technologies. Even so, acceleration is a distinct possibility, he said.

"As the potential demonstrates itself, we'll be quick to capitalize on it," he said. "Assuming it turns out to be something important, we'll pursue it pretty swiftly."

George K. Muellner, a retired USAF lieutenant general and now president of Boeing Phantom Works, which is building the X-45, was also involved when the initial SEAD UCAV concept originated in the Pentagon. Muellner was a key acquisition official during the 1990s, serving as head of what became the Joint Strike Fighter Program and as the principal deputy for Air Force acquisition at the Pentagon.

"When the Air Force and DARPA got together, the SEAD mission was really a pressing concern," Muellner said. "We were boring a lot of holes in the sky over north and south Iraq," flying defense-suppression missions as part of Operations Northern and Southern Watch.

Loitering

USAF wanted a long-loiter-time capability that would relieve pilots of having to spend hours flying around waiting for something to happen, as well as reduce the cost of maintaining the aerial blockade and avoid the potential embarrassment of having a pilot shot down over Iraq.

Furthermore, the 1999 war over Kosovo demonstrated a significant shortage of available defense-suppression capability. There was never enough to go around. Moreover, the soft kill nature of SEAD assets meant that enemy air defenses were a threat throughout the conflict. In the 1991 Gulf War, Muellner said, because mobile air defenses had been spotted in many different places but not destroyed, "the database kept getting larger and larger. ... There were actually more [potential] threats out there [at the end] than there were in the beginning."

These considerations spurred USAF and DARPA to join forces, Muellner said.

The X-45 concept calls for development of an airplane that would cost half as much as an F-16 and be 75 percent cheaper to operate. It could sit dormantly in a box, wings removed, for years, then be unpacked, reassembled and made mission-ready within a half-hour. The boxes themselves would be easily airlifted-six could fit in a C-17, 12 in a C-5-so they could either be part of a rapid deployment package or simply wait in a storage facility overseas as pre-positioned war materiel.

The boxes are not packing crates but climate-controlled containers wired to the aircraft, monitoring its health. Many such containers could be stored in a warehouse, with a single person monitoring them.

Far from the popular misconception of something like a toy radio-controlled airplane, the X-45 is a jet-powered aircraft 27 feet long and nearly 34 feet wide in wingspan. Its weapon bay will be able to accommodate two tons' worth of ordnance.

The vehicle itself will have an Electronic Warfare suite comparable to that on the old F-4G Wild Weasel airplane for roughly locating and then precisely homing in on enemy radar emitters. This will be coupled with a synthetic aperture radar that will map the target area and look for telltale signs of a surface-to-air missile setup or anything else it was programmed to find. The radar will allow precise coordinates to be obtained which will allow a GPS-aided munition to destroy it with high precision.

Jumper said that Miniature Air-Launched Decoys, or MALDs, will sweep into enemy territory, causing enemy radar operators to switch on. The stealthy UCAVs will be waiting above and will attack the radars instantly when they begin emitting. The UCAV could be the "continuous suppression" platform, a "loitering EW [Electronic Warfare] killer."

The Essential Human

En route to the target, the UCAVs would be run by an operator back at base. His principal job would be to monitor the health and progress of the four or more vehicles in flight and give them permission to fire once they found and identified their targets. The operator would not "fly" the aircraft. The robotic machines would carry out their mission, takeoff to landing, autonomously. Right now, it is assumed that a "human in the loop" will be needed to consent to weapons release, at least until UCAVs establish a track record of reliability in finding the right targets and employing weapons properly.

Initially, UCAVs will use current inventory weapons, such as the High-speed Anti-Radiation Missile and the satellite-guided Joint Direct Attack Munition. However, UCAVs will be among the first to benefit from parallel development of the Small Diameter Bomb, which will have the explosive effect of a 2,000-pounder in a 250-pound munition. The Small Diameter Bomb could more than quadruple the number of targets a UCAV could hit on one mission.

Since roughly 80 percent of a traditional fighter aircraft's useful life is taken up by training sorties, the UCAV will save enormous amounts of money by staying crated up most of the time. Operators will train on the same equipment they would use in an actual mission.

"To them, they're almost unaware as to whether ... they're operating a vehicle or whether they're operating a simulation," Muellner asserted.

"For the vast part of the mission-takeoff, landing, etc.--the operator has no direct involvement" in what the UCAV does, he explained. Only at the point when the UCAV discovers a target and asks for permission to strike it will the operator get involved by confirming that the target is legitimate and approving weapons release, Muellner said. In the not-too-distant future, the machines probably will be trusted to do even that, he added.

"As the vehicle operates, those algorithms will get smarter and smarter. Those algorithms all employ neural networks that allow the system to learn. So, it will be able to better identify and 'fingerprint' sensors, so that when it gets a ground emitter, it will recognize whether it's one that it's 'seen' before or a new one."

Leaf dissented, however, saying it will take quite some time to develop sufficient confidence in armed robots that they could be trusted to undertake lethal action on their own. For most of this decade, he noted, UCAVs will have to fly within US airspace, which is "very full" of civilian air traffic.

Building a Database

A bonus byproduct of the UCAV being in the thick of enemy defenses will be its ability to contribute to a database of threats that will build as an air campaign goes on, Muellner said. The design team is assuming the UCAV will be used for reconnaissance as well, even though that is not a primary function of the program. Its array of sensors will generate a wealth of information.

"You've got a wideband distribution network as part of the basic architecture," as well as "multiple channels so that you've got redundancy, plus you've got these sensors in a forward location," Muellner said.

Again, Leaf was not so sure. The information collected by the UCAV will be "in a useful format to the UCAV" but may not be so useful to manned aircraft, which will continue to depend on voice communication for much information-sharing during missions. UCAVs, he noted, don't talk. However, some sort of "machine interface" might be created to make UCAV information widely available, Leaf acknowledged.

Even so, Leaf asked, "Do you want to use bandwidth to continuously transmit data, or do you want to have a methodology that selectively shares data or is facilitated by an operator in the loop? That is what has to be sorted out. The fact that it has information doesn't necessarily mean that particular information has to be shared." Leaf added that USAF is constantly working on conserving bandwidth.

The UCAV will not be a disposable system; it will be built to last for "many, many missions," Muellner asserted. However, the cost of the system is such--and the technology benign enough--that it would not be a crippling loss to have one shot down, he maintained. An enemy that captured the remains of a UCAV would find little of value, since the true engineering marvel of the craft is its software, which would be destroyed.

He noted that UCAVs will likely follow the "spiral development" scheme, in which basic versions of a system are fielded, and improvements added consistently, rather than waiting for the full-up capability in the first deployment.

"We have offered the Air Force an incremental fielding approach, which John Jumper has renamed an 'effects-based fielding approach,' " Muellner said. Such a scheme would "allow the Air Force to get platforms in place earlier, the first one being a SEAD platform, then we'd move to the next block, where you'd bring on a strike capability, ... and then finally into the directed energy [block]."

Muellner explained that Boeing has discussed with the Air Force the prospect of employing directed-energy weapons on the UCAV, since there will be ample power generation capability on board. These could be lasers or high-powered microwaves, which could be used to "cook" the sensitive electronics of ground-based systems.

Whether it be launch vehicles or radars and command and control, Muellner said, the likely targets are all "heavily electronics-dependent, and obviously, high-powered microwaves can do a lot of damage to those types of systems."

In an Instant

The main benefit of lasers and microwaves is "instantaneous time-of-flight [and] high-speed suppression," he added.

Muellner also said Boeing thought out the X-45 very carefully, and even though it is an "X-plane," an operational version would be highly similar to the testbed. That way, the transition from a technology demonstration to a fielded capability could be swift.

Leaf could not anticipate how much of the Air Force might adopt UCAVs as the prime fighting vehicle. However, he acknowledged that the F-16 fleet will begin retiring in large numbers in the middle of this decade-about four years before the Air Force will receive replacements in the form of the Joint Strike Fighter--and this fact is "certainly a consideration" when evaluating the potential of UCAVs to supplement--or replace--parts of the manned fleet.

Deptula cautioned that the enthusiasm for UCAVs should be tempered with a critical eye toward the art of the possible. He said some "tend to fall into a trap" concerning UAVs and that "they tend to think only ... about putting bombs on target." Moreover, UAV enthusiasts are striving to assume certain tough roles. "We can't quite foresee replacing the human element in the aircraft," he said.

Deptula doubts that computer-brained UCAVs could compete with pilots in situations like dogfights where "you need to rapidly assimilate information that's acquired on the spot."

Still, Deptula bridled at the idea that the Air Force isn't interested in UCAVs because of the white scarf mentality.

"Where are they getting that from?" he fumed. "That's nuts. ... I don't see any institutional resistance. Quite the opposite. I see folks who think there's a lot of potential there and that we need to exploit that."

He added, "You don't hear anybody talking about eschewing space platforms because there's not a guy flying them."

Leaf would not say whether UCAVs are a leading candidate in the ongoing analysis of alternatives for a replacement tactical jamming platform to fill in behind the EA-6B Prowler. However, he emphasized that no system with the potential to do the mission effectively and efficiently has been "ruled out."

Given their size, inexpensiveness, responsiveness, and substantial onboard generating power, UCAVs could be a "natural" for the EW role, a senior USAF official said.

Deptula said UCAVs have not been "gamed" in the current QDR process because they will not appear before the end of the Future Years Defense Plan. However, notional gaming has been done with UCAVs in the 2017 period. UCAVs having been assigned "certain attributes" of capability they could reasonably expect to have by then.

He reported that in such games, UCAVs "make a difference."

"They are wonderful things and they do hold a lot of promise," said Deptula.

"But there's a whole lot we have yet to develop in order to figure out the answers to questions like, ... how many?"

Whether the UCAV turns out to be "supplementary" to the manned aircraft fleet, or even replaces a "significant" chunk of it, "the transition ... will be evolutionary," Deptula said. In Jumper's opinion, the UCAV "has great utility, especially in the defense suppression role, and we are working on the concept of operations of how this thing will be used, so that it doesn't compete for very scarce airlift resources."

He said USAF was looking at whether UCAVs might self-deploy to a war zone, employing "auto-refueling capability." Nothing will be decided, though, until after USAF is satisfied that the technology works, "when we get the thing developed and we see what we have."

The Navy Approach The Navy is under way with its own UCAV projects-also in collaboration with DARPA--that are significantly different in scope. The Navy's main requirement is for an aircraft of about the same size as the Air Force's and stealthy as well. It will have to fly further, however, going deep inland to serve as the Navy's eyes ashore, looking for theater ballistic missiles and air defenses that could threaten carrier strike aircraft. It will also have the ability to laser-designate a target for other aircraft or itself, and it will also carry its own weapons, internally. Boeing and Northrop Grumman are vying for the project, which will run about two years behind the Air Force effort. Northrop Grumman's proposal, called Pegasus, is diamond shaped, while Boeing's resembles a "scaled-down B-2," according to George Muellner, head of Boeing's Phantom Works. The Navy UCAV will need to use the carrier's catapult for launch and catch the arresting wire to recover. This technology is already in hand, according to Randy Secor, Northrop Grumman Pegasus program manager. "What we have done recently with ... the Air Force and the Navy is something called SR­GPS, which is Shipboard Relative Global Positioning System," he said. The system links GPS receivers on a landing aircraft with receivers on the ship and transmits the ship's rolling and heaving motion instantaneously to the landing aircraft. The aircraft knows from second to second exactly where the ship is and whether it's rising, falling, or rolling left or right. The autopilot translates this movement and adjusts the flight path accordingly. The synchronization is "almost down to the centimeter," he reported. Tests were done this spring where an SR­GPS­fitted F/A-18 landed itself aboard USS Theodore Roosevelt. There was a pilot onboard, just in case, but the Hornet caught the third wire of the arresting system-as good as the hottest Navy fighter jock. The Navy UCAV will be smaller than an F/A-18 but will not sit in crates. Secor said the Navy would not take on a system that would take up precious carrier deck space unless it could "use it every day." Pegasus will be stealthy, but Secor said it's still a challenge finding stealth coatings and materials that can hold up in a salty, humid environment. "I would not say we have that solved today," he acknowledged. "But we don't see that as insurmountable." The Navy UCAV will have to carry the full range of naval air-launched ordnance, and the craft must meet stringent cost requirements: one-third the purchase cost and one-half the operating cost of an F/A-18C. If the system proves useful and compatible with carrier operations, a development program could be launched in 2008 and an initial operating capability achieved in 2012. The Navy is also under way with a Vertical Takeoff Unmanned Aerial Vehicle, or VT­UAV, which has progressed to engineering and manufacturing development, the last stage before production. Though intended for surveillance, the VT­UAV, being developed by Northrop Grumman, might carry small munitions. Boeing is also pursuing a vertical takeoff UAV, for the Marine Corps, called the Dragonfly. This aircraft is a canard rotor-wing, in which the rotor blades perform as helicopter blades for takeoff and landing, but which convert to locked wings in high-speed flight. Muellner said Boeing sees the aircraft as a natural escort for the V-22, since it can match the Osprey's speed and vertical takeoff and landing capability. The Dragonfly would operate from large- or small-deck carriers or the back of destroyer-sized ships.