Predator (4 page)

*   *   *

The epiphany Karem experienced while working on his antiradar decoy for the Israeli Air Force was the primary reason he turned his attention to drones, but it was not the only one. In the late 1970s, Karem believed that RPVs were one type of aircraft a lone inventor could still develop in a garage; he was also convinced that he could build something infinitely better than the RPVs created by the big aerospace companies. Big aerospace companies didn't build many pilotless aircraft because their primary customer, the military, remained largely uninterested in RPVs other than as target drones. If military commanders thought about unmanned aircraft at all, they usually thought of them as fragile toys. Drones were a niche technology, and always had been.

The U.S. military first tried to build pilotless aircraft during World War I, when the Army experimented with a couple of “aerial torpedoes.” The one that came closest to being produced was the Kettering Bug, a tiny biplane designed to take off from rails and deliver a two-hundred-pound warhead fifty miles away, guided by a complicated autopilot system that involved a barometer and a gyroscope. The Bug never really worked as planned, until a test a few days before November 11, 1918, the day World War I ended. On Armistice Day, the War Department cancelled the project. For a few years afterward, the Army and Navy funded research into radio control of aircraft, but absent a war both services quickly lost interest.

In the 1930s, the idea of using radio signals to fly airplanes remotely was pursued by Hollywood actor Reginald Denny. The British-born Denny, who had served as a Royal Flying Corps aerial gunner in World War I, was a pilot and model aircraft enthusiast who owned a hobby shop on Hollywood Boulevard. Denny first created a company to produce radio-controlled model aircraft, then expanded into producing slightly larger monoplanes as radio-controlled target drones for the U.S. Army, which used them to train antiaircraft gunners. During World War II, the Army bought more than fifteen thousand of these balsa-and-plywood target drones from Denny's Radioplane Company, a fact Denny touted to an actor friend whose Army Air Forces job was to publicize Hollywood's contributions to the war effort. When future U.S. president Captain Ronald Reagan sent Private David Connor to the Radioplane Company to photograph its drone assembly line for the Army's
Yank
magazine, Connor discovered a girl so fetching he returned later to take more photos of her. Eventually he showed the photos to movie studio contacts, and after the war Norma Jean Dougherty left Radioplane for a legendary Hollywood career as Marilyn Monroe.

Another World War II attempt to use radio-controlled aircraft ended far less happily, and probably changed U.S. political history. In 1944 the Army Air Forces undertook a secret plan to attack sites in occupied France and Holland that the Germans were using to launch V-1 “buzz bombs” and V-2 ballistic missiles, weapons that were terrorizing London and inflicting heavy civilian casualties. Under the Army's Project Aphrodite, twenty thousand pounds of high explosives were packed into much-used B-17 bombers, which were rigged to be flown by remote control, then crashed into targets by a “mother ship” flying much higher. Getting such a flying bomb airborne by radio was problematical, so a crew of two airmen was needed to take off in the modified bomber, designated a BQ-7, to arm the explosives and engage an autopilot to turn control over to the accompanying mother ship, then bail out over England. On August 4, 1944, four BQ-7s were launched. One exploded and killed its crew before reaching the English Channel; none of the others reached its target.

Eight days later, under the code name Project Anvil, the U.S. Navy tried the same technique to attack a site at Mimoyecques, France, near Calais, where the Germans were building guns to bombard England with giant shells. The Navy's explosives-packed PB4Y-1 Liberator, a B-24 modified for radio control, exploded shortly after takeoff, eight miles southeast of Halesworth Airfield, in Suffolk. No trace was ever found of the radio control expert on board, Lieutenant Wilford John Willy, nor of the pilot, Lieutenant Joseph P. Kennedy Jr., older brother of John F. Kennedy and the son who had long been groomed by their father to run for president. Kennedy and Willy were posthumously awarded the Navy Cross.

Radioplane continued to sell the Army its propeller-driven remote-control target drones after the war, and in 1955 the company added a film camera to one, creating the world's first unmanned reconnaissance aircraft. The Army bought hundreds. The Air Force, however, was generally uninterested in planes without pilots inside—until May 1, 1960, when the Soviet Union shot down a high-altitude U-2 reconnaissance jet flown over its territory by CIA pilot Francis Gary Powers.

Eight days after Powers was downed and captured, the Air Force secretly awarded a contract to Ryan Aeronautical Company to adapt its jet-powered Q-2 Firebee target drones for photo reconnaissance. The project was blocked a few weeks later, just as flight tests were being conducted, by Pentagon bureaucrats with other priorities and Air Force leaders more eager to spend their reconnaissance money on a sexy new manned spy plane called the SR-71. The SR-71 was going to fly at Mach 3—three times the speed of sound—and at altitudes no Soviet missiles could reach. The Air Force got interested in Ryan's reconnaissance drone again, though, when, on October 27, 1962, one day before the end of what became known as the Cuban Missile Crisis, another U-2 was shot down over Cuba and its pilot killed.

Two years later, the United States sent troops to South Vietnam to fight Communist insurgents, and over the next eight years the Air Force modified and deployed more than twenty derivatives of Ryan's jet-powered target drones for photo reconnaissance. The drones were also used as decoys to fool North Vietnamese air defenses, to drop propaganda leaflets, and to carry sensors able to eavesdrop on enemy communications. These Firebees, Fireflys, and Lightning Bugs, as they were variously designated, flew 3,435 missions over eight years.

The technology was hardly convenient, however: the miniature jets could neither take off nor land like regular aircraft. They had to be launched from under the wing of a C-130 transport plane to fly their missions while being remotely controlled from a ground station transmitting radio signals susceptible to interference from other radio traffic. Early versions deployed a parachute to land over South Vietnam—sometimes in a rice paddy, sometimes in the jungle, sometimes in the ocean off Da Nang, sometimes in hostile territory. To avoid damage from hard landings, saltwater contamination, or just losing a drone, later versions were plucked from midair by a beefy CH-3C helicopter after their parachute opened. Manipulating two twenty-foot-long hydraulically operated poles and an array of three hooks, the helicopter crew would snag the parachute's cords and, with a winch that fed a thousand feet of steel cable out a reinforced hole in the chopper's floor, reel the two-thousand-pound drone into a position about twenty feet underneath the helicopter for delivery to a recovery zone. The drone's still-photo film cartridge then had to be flown out of Vietnam for development and analysis, with photos of interest flown back later, a process that sometimes took days. Often there was no recovery to perform and no film to develop. On average, such drones flew fewer than four missions apiece. More than half of these RPVs, 544, were shot down or crashed.

Despite these experiments, no large constituency for drones ever developed within the military. The reasons were largely cultural. The pilots who ran the Air Force usually didn't pay much attention to aircraft that flew without a member of their proud fraternity inside. The Army, committed to helicopters after Vietnam, wasn't particularly interested in any sort of airplane, so there was no natural home for drones in that service. The Navy experimented with its helicopter RPV and some Ryan drones during Vietnam, but ship captains never liked the idea of aircraft with flammable gas that flew without a pilot inside to handle an emergency landing on their decks. Besides, once the Vietnam War ended and defense spending plunged, there were more pressing and promising uses for available funds. Throughout this period, the only sectors of the government that remained truly interested in drones were the CIA and a super-secret intelligence agency called the National Reconnaissance Office, which used some exotic RPVs to spy on the Soviet Union and China during the Cold War. But even their interest blew hot and cold; among the cloak-and-dagger set and their politician bosses, spy satellites were seen as more promising.

When Karem arrived in America, he saw a vast new frontier beckoning. As he surveyed it, he was sure he could make his mark in his new country, and maybe a fortune as well. First, though, he had to invent a better drone, and find a customer willing to buy it.

*   *   *

From the time he joined the Aero Club in Israel as a boy, Karem had read every book, magazine and newspaper article, and every scholarly paper he could find that had to do with flight. For nearly every airplane or helicopter ever built anywhere in the world, he could reel off from memory not just the numerical designations and names but also figures and facts, including who had designed the aircraft. He had also studied the myriad methods of flight found in nature, whose aerial creatures employ an amazing array of airfoils (wing shapes) to travel through the fluid known as air. They do so in a wide variety of ways. Songbirds flap and flit on wings that taper sharply. Hummingbirds hover and dart by beating their tiny wings dozens of times a second. The housefly's flat, wide wings are less airfoil than paddle, leading some scientists to conclude that, given a fly's typical weight of twelve milligrams or less, the air is so viscous that the insect less flies in it than swims.

Nature's best soarer is the fish-eating albatross, a seagoing bird that glides over the ocean for hours at a time, floating on outstretched wings whose span is roughly twenty times longer than their chord, meaning their width front to back. This “high-aspect-ratio wing” is the shape favored by free-flight modelers. When Karem set out to build his better drone, he took the albatross as his model. More schooled in ornithology than in eighteenth-century English poetry—the language was his fourth, after all—Karem called his new drone the Albatross. At the time, he was unaware that in English the name had been synonymous with “burden” ever since 1798, when Samuel Taylor Coleridge published his epic poem “The Rime of the Ancient Mariner.” Karem's Albatross, however, would prove to be not a burden but a boon.

Thoroughly schooled in the hapless history of unmanned aircraft, Karem believed RPVs had largely failed to catch on for two reasons: most could fly for only a couple of hours at a time and most fell out of the sky at rates that would raise alarms if anyone were inside them. He blamed those flaws on the fact that most RPVs had been developed either by modelers accustomed to making toys that were cheap to build and replace or by aerospace corporations whose best people worked on more lucrative products and whose unmanned aircraft were designed, like target drones, to be expendable, not least because their customers, the military, expected no better.

Karem took an entirely different approach. At IAI, he had always encouraged his staff to be innovative because, as he liked to say, “The customer doesn't really know what he wants.” What Karem meant was that military officers couldn't possibly know enough about the latest technologies to understand what was feasible. At weekly meetings, he had instructed his IAI engineers to figure out what the customer wanted, then try to design something providing three to five times as much capability. Now working out of his garage in Los Angeles, Karem was determined to design his Albatross well enough to show potential customers that a drone could stay airborne not just for hours but for days at a time. He was also convinced that he could build his RPVs with the same rigor and reliability required of the fighter planes he worked on in Israel.

By a stroke of good fortune, one of the first people to befriend Karem in Los Angeles was Ira Kuhn, a physicist consultant to the Defense Advanced Research Projects Agency, known by the acronym DARPA. Then, as now, DARPA existed to fund farsighted and even far-fetched ideas to help the military stay on the cutting edge of technology. Kuhn and Karem met at a tiny Santa Monica company called Developmental Sciences Inc., where Karem had gone to work in 1977 to gain a foothold in the U.S. aerospace industry. Karem was helping the two engineer owners develop a couple of RPVs, including one for DARPA whose progress Kuhn was monitoring for the agency. When Karem quit Developmental Sciences in 1980—storming out one day after the owners refused to make him a full partner—and went to work full-time in his Hacienda Heights garage, Kuhn persuaded DARPA to fund development of Karem's Albatross. Fearing Defense Department auditors would object to funding an aircraft being developed in a garage, however, DARPA gave Kuhn's consulting company a $350,000 contract to finance Karem's work. “We were DARPA's conduit to get it to him,” Kuhn explained years later. Karem's work in Israel had such a good reputation that, even if Karem
was
working out of a garage, DARPA Director Robert Fossum figured his Albatross project was a good bet.

The Albatross wasn't meant to be an operational RPV, just a technology demonstrator. It wasn't much bigger than a model, and was made mostly of mahogany plywood, spruce, urethane foam, and fiberglass shaped in molds Karem had fabricated himself. The diameter of the fuselage's bullet-shaped nose measured 300 millimeters, or 11.8 inches, the same as the unvarying chord of its 15-foot wing. The wing sat atop the fuselage, midway between the nose and tail; at its rear end, a small vertical stabilizer 18 inches tall and half as wide pointed straight up. Two larger tail fins, or horizontal stabilizers, extended down in an inverted V, a configuration Karem chose so those appendages could serve as skids and keep the pusher propeller on the tail from hitting the ground during runway landings. The propeller, meanwhile, was powered by a two-stroke, single-cylinder McCulloch 101 go-kart engine. Radio-controlled, the Albatross could take off and land like a regular airplane using detachable landing gear and a nose wheel. It also had a parachute for emergency landings. Most notably, Karem's little drone weighed 105 pounds when totally empty but could carry 95 pounds of fuel, an uncommonly high “fuel fraction” of 47.5 percent—a key feature of aircraft endurance.