Home : Three Services :

High-risk, High-return Experimentation

You’re in a cave overlooking a canyon, lying in wait for a squad of approaching U.S. soldiers. Kalashnikov locked and loaded, you are primed for the ambush. You think you have the upper hand. But you’re wrong.

Eleven months ago, a humongous blimplike eyeball lifted off from a distant U.S. base; it’s been hovering, invisible, 70,000 feet above you ever since. It’s three times the size of the Goodyear, and inside this floating death star is an antenna nearly as long as the Statue of Liberty. More powerful than even the most sophisticated satellite, the airship has sensors that can track the movement of individual soldiers on a battlefield. It’s been following your positions for days, along with those of thousands of other potential targets.

Six thousand miles away, at CENTCOM’s base in Tampa, a technician looks up from his jelly donut and notices you and your squad. While sipping his coffee, he orders the deployment of one of Lockheed Martin’s morphing “hunter killers,” unmanned hawklike predators with outstretched wings and blunted beaks. Approaching your position, it folds its wings to about half their former size, then turns its nose straight down, diving like an osprey going for a fish. Just before impact, it pulls up and releases a laser-guided bomb on your head. You’ve just become a red stain on a rock, and donut boy goes back to browsing Maxim.

The blimplike eye in the sky is called ISIS (short for Integrated Sensor Is Structure). Neither it nor the hawk-drone is on the prowl just yet, but they’re among the more promising projects currently coming out of the Pentagon’s Defense Advanced Research Projects Agency, or DARPA. In March the agency made headlines by sponsoring the Grand Challenge, a 142-mile road race between 14 different unmanned vehicles with a $1 million purse. The idea, of course, is eventually to create a fully automated mechanized cavalry, capable of defeating the enemy without risk to American soldiers. In the end, none of the vehicles got more than about seven miles past the starting line in Barstow, California, but the race drew big weekend crowds and national TV coverage. What few realized, however, was that earlier that week, even wilder high-tech projects and killing machines were on display in Anaheim, California, where DARPA held its annual conference, a block from Disneyland.

It’s not such an absurd juxtaposition. There is a definite layer of fantasy to what DARPA does. Created in 1958, in the days following the Soviets’ launch of Sputnik, DARPA’s always been about high-risk, high-return experimentation, about bringing together the best and the brightest and saying, “This sounds impossible…but what if?” It just so happens that the “what if” often involves creating impersonal, Terminator-like machines.

At a high-caliber Pentagon trade show, where military contractors unveil the next generation of death machines, among the more exotic projects on display at the Marriott Hotel that weekend: tiny robots programmed to swarm like killer bees, miniature fuel cells powered by an ant pheromone, satellites powered by water rockets, and a handheld language-busting computer not unlike Star Trek’s universal translator.

The coolest exhibit by far was an “exoskeleton” that can turn a soldier into a modern-day Hercules. The contraption consists of two mechanical leg braces, a power unit, and a backpacklike frame. More than 40 sensors and hydraulic mechanisms mimic the human nervous system, functioning like muscle. If perfected, the device could eventually allow supersoldiers—and later, civilians—to heft giant loads.

Few of the projects had gone much further than the drawing board, but there was a palpable air of almost whimsical fun on the convention floor. Raytheon vice presidents in finely tailored silk suits stood by exhibits, talking robotics with geeks from Carnegie Mellon with Leatherman tools and Velcro-sheathed flashlights strapped to their blue jeans. Squared-away colonels in immaculately pressed greens discussed missile kill radius with middle-aged suburban garage inventors. At one end of the conference hall, Srikanth Saripalli, a 25-year-old graduate student at USC, was showing off a miniature remote-controlled helicopter that can drop tiny sensors in enemy territory to monitor troop movements. “We are trying to learn how to fly through urban canyons,” he was explaining when a three-foot-long robot tank appeared out of nowhere and, with a horrific grinding noise, ran over his extension cord. “Sorry, sorry, sorry!” shouted a pale, nervous-looking man with a remote control in his hand.

DARPA’s failure rate is said to hover at about 85 to 90 percent, raising the question of whether it’s all just a giant subsidy for nerds. But the possibility of failure—indeed the open embrace of its probability—is part of what has allowed the agency to survive sublimely idiotic aborted projects (robotic elephant) and pursuits (telepathic spies). Failure and risk are built into DARPA’s mandate, and that is oddly liberating. “We generally try and set milestones to evaluate the progress of a project at about 12 to 18 months,” says Stephen Welby, deputy director of the Orwellian-sounding Information Exploitation Office. “Sometimes we discover it’s a great idea but it’s too hard. If it’s not paying off, we can rapidly shift and look for alternatives.”

When DARPA scores, it often scores big. This is the agency that created the Internet. Their technology enabled GPS for civilian use. They invented the stealth bomber and the computer mouse. “We have 160 technical folks out continually talking to people,” Welby says. “They’re talking to military folks, they’re living with guys in the field, they’re out at industrial labs, they’re talking to guys in garages. They come up with nutty, far-out ideas and tell us how they’re going to work. We give them the resources.”

And thanks to the war, the mother of military invention, DARPA is currently enjoying more resources than it’s ever had, with this year’s budget reaching a record $2.8 billion.

On the convention floor, it wasn’t hard to find projects with potential applications in the Middle East. There were ideas for defensive systems that can counter individual RPG rounds, detect suicide bombers in a crowded area, and neutralize roadside bombs. But in both Afghanistan and Operation Iraqi Freedom, the revolution was in information technology—combining radar systems on ships, data from satellites, and imagery from drones into a 3-D view of the battlefield for commanders; now the challenge is bringing that same sensory capability to a single dismounted soldier. Bounding onto the exhibition stage this past March to synthesizer riffs, a portly, mustached man named Larry Corey laid out the basis for this lethal vision in Hollywood terms. “Think of Harry Potter,” he said. “Harry and his friends are sneaking around on the grounds of the Hogwart School for Witchcraft and Wizardry. Harry discovers a magic map that shows the position and identity of everyone on the school grounds. Today Harry’s map is fiction. But imagine if we could give it to our Special Operations forces and commandos?”

Is such a map possible soon? Depends on how much those little wizards can stomach. But imagine dropping 100 tiny sensors in Falluja with Saripalli’s helicopter. Feed their data into a computer, then combine it with a night-vision feed from the goggles of a platoon of soldiers close to the same scene. Throw in the ISIS angle and, with proper software and a special helmet, commandos have a virtual, real-time representation of the battlefield. They can then sneak up on and blow the heads off enemy combatants like playing a video game…with minimal risk.

“I challenge you to help make this vision a reality,” Corey exhorted the crowd.

Will this, and DARPA’s other new death-dealing gizmos, become real? Who knows? We just know Harry Potter’s Hogwart Slaughter would make a great Disneyland ride. DARPA’s mission is to maintain the technological superiority of the U.S. military and prevent technological surprise from harming our national security by sponsoring revolutionary, high-payoff research that bridges the gap between fundamental discoveries and their military use.

DARPA's mission has been to assure that the U.S. maintains a lead in applying state-of-the-art technology for military capabilities and to prevent technological surprise from her adversaries. The DARPA organization was as unique as its role, reporting directly to the Secretary of Defense and operating in coordination with, but completely independent of, the military research and development (R&D) establishment. Strong support from the senior DoD management has always been essential since DARPA was designed to be an anathema to the conventional military and R&D structure and, in fact, to be a deliberate counterpoint to traditional thinking and approaches.

The Department of Defense's senior management, seeing the value of an agile, forward-looking R&D group unconstrained by conventional thinking and able to investigate ideas and approaches that the traditional R&D community finds too outlandish or risky, has consistently protected the independence of DARPA. Failure to keep the bureaucracy at bay would have doomed the value of DARPA and this has been consistently recognized over the years. The freedom to act quickly and decisively with high-quality people has paid handsome dividends for DoD in terms of revolutionary military capabilities.

DARPA's original operating philosophy has changed over the years in only three ways - its relationships with the commercial marketplace, its business practices, and its emphasis on joint systems. First, the DoD has gone from dominating the market in such areas as microelectronics, computing and network communications, each of which was driven by DARPA in past years, to the current situation where the DoD is able to somewhat influence the directions of a much-larger-than-DoD market. DARPA has played one of the key roles in assuring that DoD's long-term interests are served in this new situation.

Second, in the past decade, DARPA has pioneered revolutionary R&D business practices reform. With the support of the Congress and DoD senior management, DARPA has led the way in adopting commercial practices and innovative contracting arrangements. Congress provided the authority for "Other Transactions" and "Section 845" agreements to DARPA on an experimental basis, and, because of DARPA's success, has now conveyed the same authorities to the rest of DoD. Third, since the Goldwater-Nicholls Act, DARPA has focused considerable attention on solutions to joint-Service systems and problems.

DARPA's ability to adapt rapidly to changing environments and to seek and embrace opportunities in both technology and in processes, while maintaining the historically proven principles of the Agency, makes DARPA the crown jewel in Defense R&D and a unique R&D organization in the world.

DARPA’s original mission, was to prevent technological surprise. This mission has evolved over time. Today, DARPA’s mission is to prevent technological surprise for us and to create technological surprise for our adversaries. DARPA’s mission implies one imperative for the Agency: radical innovation for national security. DARPA’s business processes reflect this in a straightforward way: bring in expert, entrepreneurial program managers; empower them; protect them from red tape; and quickly make decisions about starting, continuing, or stopping research projects.

Adam Piore. The War Wizards. . July 2004.

top of page

back a page