Updated

A government office is sponsoring a robot race through the Mojave Desert with a $1 million cash prize, trying to build lasers powerful enough to down missiles and studying the feet of geckos. And it calls its best employees "freewheeling zealots."

DARPA (search) - the Pentagon's Defense Advanced Research Projects Agency - is unlike any other federal agency.

New assignments in terrorism and high-powered computer software have brought unwanted attention and controversy to the office of just 240 employees. Before that, it had quietly transformed not only the U.S. military, with stealth aircraft, Predator drones and precision bombs, but also American society, with the Internet and global positioning satellites.

The man whose work generated much of the recent publicity, former Iran-Contra figure and retired Adm. John Poindexter (search), left the agency Friday after 20 months.

This week, House-Senate conferees will debate a Senate-passed provision to halt spending on his Terrorism Information Awareness (search) project, which was designed to search the commercial transactions and personal records of millions of Americans to detect by computer the telltale patterns of activity by would-be terrorists.

Born in 1958 when the Soviet Union beat the United States into space with the Sputnik satellite, the agency is supposed to prevent technological surprises from U.S. adversaries. It long ago handed off ballistic missile work but still focuses on research that is over the horizon in time and space.

"We look at what future commanders might want," agency director Tony Tether says, "changing people's minds about what is technologically possible." DARPA investigates ideas "the traditional R&D community finds too outlandish or risky," its web site boasts.

"If they aren't failing more than 50 percent of the time, they are probably too conservative," says Steven Aftergood of the Federation of American Scientists (search).

Washington University's Ron Indeck, an engineering professor who's worked with DARPA, says, "They go for very high-risk, high-payoff projects."

And the Poindexter effort was high-risk in every way - as well as a classic illustration of how DARPA works.

With special flexibility to hire and pay competitive salaries, DARPA mines the defense contractors, universities and military offices that perform all its research. "They drag in people as program managers who can have the greatest impact as fast as possible," Indeck said. "Nobody stays more than three to five years."

The agency concedes "the best DARPA program managers have always been freewheeling zealots in pursuit of their goals."

Still, many were surprised DARPA dared bring in Poindexter. Forced out as President Reagan's national security adviser, Poindexter had been convicted of lying to Congress in the Iran-Contra (search) scandal, though the convictions were overturned on appeal.

Tether lured Poindexter from Syntek Technologies to head a new Information Awareness Office to develop the admiral's ideas for thwarting terrorists. DARPA paid Poindexter $138,200 a year, the same as Tether, and close to his $147,182 Syntek salary.

Poindexter's project was quickly denounced by privacy advocates from the political left and right, and Congress began restricting it.

His plan for a futures market that let investors bet on Mideast political developments was based on markets that had predicted election outcomes and movie sales, but a contractor outlined plans to sell futures contracts on assassinations and coups.

That created a political firestorm. Sen. Ron Wyden, D-Ore., denounced it as "a federal betting parlor on atrocities and terrorism," and the Pentagon killed it in 24 hours.

Two other projects from DARPA's $2.69 billion budget, highlighted in Associated Press stories earlier this year, drew criticism:

-- Combat Zones That See (search) would allow soldiers using computers and small cameras to track and record the movement of every vehicle in a city. Some feared this urban combat tool would be sold to domestic police departments.

-- LifeLog (search) would record everything someone saw, heard, did, felt, said or read and use software to automatically organize those events into stories. Some feared the story-drafting software could be applied to data beyond the user's control and end up mistakenly constructing suspicious stories.

"The agency fell short in failing to appreciate the magnitude of public concern over privacy issues," Aftergood said. DARPA has always favored "elitist research, without a lot of questions asked at the beginning," Princeton University historian Michael Mahoney said. "A wider peer review might not be a bad idea."

But these are small parts of an agency accustomed to more praise than criticism.

Next March, DARPA hopes to award $1 million cash for the first vehicle to cover 300 miles of unpaved road, Mojave Desert canyons, snow-covered rocks, and flowing streams between Los Angeles and Las Vegas in less than 10 hours. No human aboard, no remote control, no tire changes, no refueling.

DARPA is trying to reach beyond its traditional researchers and may succeed. For instance, one team entered in the Mojave race is headed by Travis M. Thul, a sophomore at Fox Valley Technical College in Appleton, Wis. He's using about $150,000 in equipment donated by Wisconsin and Tennessee companies that have never had a DARPA contract.

"We don't have the money of other entrants," Thul said. "But we have get up and go, some good people and good software engineering. It's not that hard. We'd like to show them we're not hillbillies with shotguns."

Having produced the unmanned Predator drone used in Afghanistan as well as robotic vehicles and a robotic helicopter, DARPA envisions such vehicles taking over the kind of supply mission on which Army Pfc. Jessica Lynch was ambushed and injured during the Iraq war.

Other DARPA projects are:

-- Working on high-powered fiber and liquid lasers to be mounted on vehicles, planes and ships to destroy attacking missiles.

-- Studying the feet of geckos, the tiny lizards with adhesive discs on their toes. The goal: to build small, legged robots that can climb walls.

-- Studying hibernating Arctic ground squirrels, who quickly regrow synapses that support memory, and whales and dolphins, which put half their brain to sleep while the other half surfaces them to breathe. The goal is to determine how to keep soldiers awake seven days straight without fatigue or mental impairment.

-- Seeking to create an exoskeleton for soldiers that would enable them to leap extraordinary heights

DARPA-funded researchers also have developed a system of brain sensors and mechanics that allows a monkey to move a remote robot arm just by thinking of doing it.

"We do not fully understand the implications of this work," Tether has said. "But imagine how useful and important it could be for a warfighter to use only the power of his thoughts to do things at great distances."

"Will all these technologies work? We don't know yet."