Is this the future of air combat?

For now, the UCAVs have yet to perform some very basic tasks, such as plugging into a fueling tanker in flight, sharing airspace with other military and civilian airplanes, or landing on an aircraft carrier. Until they do, the big money in the U.S. and elsewhere will remain budgeted for manned fighters, and there´s no doubt that thousands more will be built.

But it may not be long before UCAVs overcome their limitations. In early February, the Boeing pair took off from Edwards, circled over the Mojave Desert, automatically attacked a simulated missile site, and returned to their orbits. Minutes later a second missile site, unknown to the UCAVs´ computers, advertised its presence by sending simulated radar signals, and the UCAVs attacked it, too. The ground-based operator´s only job was to OK the release of weapons.

Much more testing is needed-testing that will involve half a dozen much larger prototypes, due to start flying in early 2007. Next up are three Boeing X-45Cs and Northrop Grumman´s X-47B prototypes, all being developed as part of a $4-billion program launched last year by the Defense Advanced Research Projects Agency (DARPA). The X-45C and X-47B are much bigger, to accommodate more fuel and weapons, than the four-ton X-45A. This is because DARPA program manager Michael Francis and other researchers believe that they have identified both an important advantage of the UCAV-its endurance isn´t limited by the pilot-and a mission that it can perform better than any current fighter. Northrop Grumman´s UCAV manager, Scott Winship, calls it â€the supergapâ€: the ability to loiter for hours inside enemy territory and attack moving targets, such as missile launchers. â€They can get in close and hit anything, up to 1,000 miles deep,†Winship says. â€It´s something we can´t go after today.â€

But if the UCAV is going to unleash deadly force, will it do so automatically? That´s an idea that makes people uneasy, especially when civilians are at risk. If it needs to â€ask†a human operator for permission to launch a weapon, however, the response may come too late-a fast-moving fighter jet can vanish in seconds. And Jumper has raised concerns that the UCAV might not be able to survive in combat because, as currently conceived, it has no way to detect a fighter attack and won´t be agile or fast enough to evade it. One UCAV proponent inside the industry, who prefers not to be identified, interprets this position as a way for the Air Force to justify more F-22s, as top cover for the vulnerable UCAVs.

UCAV visionaries see these issues not as deal breakers but as challenges for the next few years. One of Francis´s goals is to create software that allows a group of UCAVs to intelligently, collectively attack a target or a threat-but never twice in the same way. In action, the UCAVs would rarely operate alone but as a wolfpack of four or more aircraft, collaborating to locate targets. Only one airplane in the group would use its radar at any time, making their signals very hard to track. The aircraft would defend one another with jamming and weapons. Winship suggests that one of the airplanes would carry a battery of air-to-air missiles and engage any fighter as soon as it got off the ground. (There may well be more than one type of UCAV. Winship sees everything from a 20-ton Naval version, the size of the X-47B, to a RoboBomber as big as a B-2 bomber, capable of flying 100-hour missions. To save money, they´d use common weapons, computers and radar.)