Planes that stay aloft for five years. Vaccines that take weeks, not years, to develop. Surveillance drones that are launched into combat zones atop ballistic missiles.

All these futuristic military projects are outlined in the just-issued Strategic Plan 2009 of the Defense Advanced Research Projects Agency (DARPA), which lays out what the high-tech Pentagon division — responsible for developing the Internet, among other things — is working on.

But are expensive projects like these, with no guarantee of success, really worth developing as the Pentagon pares back its budget?

Defense Secretary Robert Gates' recently axed the Future Combat Systems program and the F-22 fighter, leading many to suspect that high-concept research and development ventures could be next to go — even as some say that would be shortsighted and might jeopardize America's technological competitiveness.

• Click here for FOXNews.com's Patents and Innovation Center.

"It comes down to who Obama appoints to run DARPA, and how much or how little they follow Tony Tether's lead," says Lisa Wright, press secretary and energy legislative assistant for Rep. Roscoe Bartlett, R-Md., who sits on the House Armed Services Committee. (Rumors have RedX CEO [and former DARPA manager] Regina Dugan as the pick.)

Tether, appointed by Defense Secretary Donald Rumsfeld in 2001, was the longest-acting director in DARPA history when he left in February, after the Obama inauguration. Contrary to tradition, he was a very hands-on manager.

"Tether ran DARPA much more like a business," says Wright. "If a program didn't produce immediate results (meaning in a year or two) he was very quick to cut it. What we don't know is if the next chief will continue this policy or revert to the earlier, more latitude-to-fail, model."

Before Tether, DARPA didn't act like other military agencies, or indeed like any typical government agency. It was built to take risks, costs and benefits be damned.

"This isn't government as usual," says Joe Katzman, editor-in-chief of Defense Industry Daily. "DARPA runs on the same model as a venture-capital firm, and uses the same metrics for success.

"If a VC firm funds 10 companies, what they're hoping for is one out of those 10 become a hit. One success is enough to make them fabulously wealthy."

One potential hit for DARPA is MIMIC, or Modular IMmune In-vitro Construct, a surrogate human immune system that could rapidly develop vaccines to combat unforeseen biological threats, such as the current swine-flu virus.

Other potential winners are Rapid Eye, which launches an unmanned aircraft into a crisis atop a ballistic missile, and Vulture, a solar-powered unmanned plane meant to stay aloft for 5 years at a time.

"During a crisis, to develop vaccines quickly, we need real human data," says William Warren, CEO of the Florida-based biotech company VaxDesign and the subcontractor who built MIMIC. "It's not just that animal testing is controversial, expensive and time-consuming. It's that a lot of drugs that work in animals don't in humans. MIMIC allows clinical trials to take place in a test tube — it sidesteps these problems."

MIMIC, which accurately predicted that the 2007 flu vaccine would work on only half the population, has already gone through DARPA's developmental Phase I and Phase II, meant to weed out weak projects.

Next month, it enters Phase III, in which a military branch comes in to help pay for final development. It's the real proof that a project is going to work, and the Army is expressing serious interest.

The military-partner requirement for Phase III acts as a check, keeping untenable pie-in-the-sky ideas from becoming bank breakers — a method that's possible because of how DARPA attacks problems.

"DARPA tries to surround real-world military issues with a number of different solutions," says Graham Warwick, senior editor with Aviation Week and Space Technology. "Not all of them have to work."

Many DARPA projects are outgrowths of recognized shortcomings in combat situations in Iraq and Afghanistan and in the war on terror in general.

"Currently, our UAVs (unmanned aerial vehicles) can stay aloft for 30 hours, but these days the military wants to be able to identify a target and track that target from the air for weeks, perhaps months," says Warwick. "Nothing we have in our arsenal can pull that off."

To this end, DARPA's funded a bevy of new ideas. One is Rapid Eye, which could be launched into a suddenly developing conflict when no spy satellites are available.

Most assume the project will be cut, due less to feasibility problems than to the political difficulties of firing a ballistic missile into an already tense situation.

Two other UAV programs are Isis, an unmanned combination surveillance blimp/massive radar system, and the aforementioned Vulture, a huge solar-powered unmanned communications plane that's meant to fly at 70,000 feet for half a decade.

Vulture illustrates why some experts feel DARPA's projects are safer from budgetary constraints than other military programs: They don't actually need to succeed to succeed.

"Nothing about Vulture is arbitrary," says Warwick. "Satellites stay aloft for 5 years, so at a basic level this program is about applying that paradigm of durability to airplanes. This alone will help us build better planes."

Tucked within Vulture are the remarkable advances in solar technology and energy storage required to keep a plane aloft for five years.

Tucson, Ariz.-based Scion Power, a Vulture contractor, is working on lithium-sulfur batteries that would be twice as strong on anything on the market. If they work, they'd not just revolutionize surveillance aircraft — they'd also ease the auto-industry move to electric vehicles.

"This is exactly why DARPA serves such a great need," says John Langford, president of Aurora Flight Sciences, another company involved with Vulture. "People think technologies go straight from the lab into production, but there are lots of very costly in-between steps.

"Without someone like DARPA stepping in, these batteries would be just too expensive for the auto industry to develop."