Brian Cooper may be the slowest driver in the solar system. Floored, his vehicle reaches a sustained velocity of a tenth of a mile per hour. That's fine with Cooper, who fears getting stuck in the bottom of a crater. On Mars you can't call a tow truck.

On a recent morning, Cooper's vehicle, the rover named Opportunity, stood parked on the downward slope of a crater the size of a football field. On his computer here at NASA's Jet Propulsion Laboratory, Cooper could see what the rover saw through its stereoscopic cameras. Toward the bottom of the crater, where the surface leveled out, the rocky ground gave way to softer soil. Trouble.

Cooper wanted to tell the rover to take this opportunity to turn the heck around and get out of the crater. But scientists wanted the rover to keep going down, deeper into the planet's geological past. They won the argument.

"I'm just one little voice of caution. Our job is to make sure the rover is safe to drive another day," Cooper said. "We never go fun-truckin'."

He's an engineer, and engineers know better than anyone the limits of their contraptions. He does a great imitation of Scotty from "Star Trek":

"It can't take it ennymore, Cap'n!"

The triumphs of two rambling rovers on Mars, as well as the success of the Cassini spacecraft that recently reached Saturn, prove that NASA still knows how to perform magic in space. But it is simultaneously a traumatized agency, struggling to maintain the dream of the Apollo era. The space shuttle fleet remains grounded as a result of the Columbia disaster. The International Space Station orbits the Earth aimlessly, an expensive contraption that remains unfinished, with two lonely astronauts laboring to keep it from falling apart.

Earlier this year, President Bush endorsed a bold plan to return astronauts to the moon and eventually send them to Mars. This "Vision for Space Exploration" would allow NASA to retire the aging shuttles and eventually put the space station out to pasture. Space buffs are excited by the prospect of getting beyond Low Earth Orbit again. But Congress hasn't rushed forward with money for a moonshot or a Mars mission. The last such ambitious onward-to-Mars proposal, the Space Exploration Initiative of the first President Bush, went nowhere when Congress saw the price tag. Even this President Bush, days after unveiling the plan in January, declined to say a word about it in his State of the Union Address. It was as though sending men to Mars was a momentary brainstorm, one of those ideas that seems great at midnight and is quickly forgotten the next morning.

There has been one dazzling feat of human spaceflight in recent months, but it was a private venture, a piloted rocket-plane that bored a hole in the sky, touched the edge of space more than 300,000 feet up and safely returned to Earth.

The government, meanwhile, hasn't yet figured out how to keep one of its signature triumphs, the Hubble Space Telescope, from falling back to Earth in a fireball. NASA doesn't want to endanger a crew of astronauts for the sake of an aging instrument that will eventually be replaced by more powerful telescopes. There is talk of a robotic mission to save the Hubble, but the whole issue has been a public relations disaster for the agency, emanating the whiff of a can't-do attitude. These were the people who could always do the impossible. They were the ones who inspired a great American cliche: If we can put a man on the moon, why can't we . . .

Even here at JPL, past disasters keep everyone from feeling overconfident. People know that spaceflight remains a difficult, experimental, chancy line of work.

"The shadow of failure is around the corner all the time," says Charles Elachi, JPL's director.

Robotic spaceflight and space-based astronomy, using a new generation of telescopes, may be the likeliest source of NASA glory for the next decade or so. But space science has never been the central purpose of NASA. NASA has always been powered by the dream that humans, and not just their robotic proxies, will directly explore the universe.

A dream that, right now, looks all too literally fantastic.

Highs and Lows

The Jet Propulsion Laboratory is not a laboratory at all, but something more like a high-tech college campus, parked against the scrubby brown hills of a sun-washed valley at the base of Mount Wilson. A visitor will want to stop by the Space Flight Operations Facility, also known as Mission Control, which is the closest thing here to something resembling the bridge on "Star Trek."

A few employees in the darkened room are stationed at computer monitors while, overhead, a screen flashes graphics of Saturn and the solar system. This is partly for show, for the entertainment of tour groups. Real science is rather uncinematic these days. The rover teams are in a different building, and Cooper works on a desktop computer, manipulating graphics, looking like any other office drone in the Information Age -- just one who happens to drive a rover on Mars.

"We see lots of slip," he says. His rover's wheels spin in the Martian soil. As he manipulates a mouse, the rover on his screen pivots, looking around that crater. Cooper can interpret features that, to an untrained eye, appear indistinct, a bunch of shadowy shapes. One of his colleagues, who drives the other Mars rover, named Spirit, has a sign on his office door that reads: "My other cars are on Mars."

When a voice comes over a loudspeaker and announces a cross-agency discussion on the reorganization of NASA, the engineers don't seem to hear it. A subsequent glimpse into the auditorium reveals only five people watching the video feed from NASA headquarters in Washington. Here, people have more exciting things to do -- like exploring Mars and Saturn.

JPL has had its own traumas. Five years ago, two robotic probes, the Mars Climate Orbiter and the Mars Polar Lander, vanished when they reached the Red Planet. When things go wrong with robotic probes, they don't tend to say goodbye. They just disappear. There's no video feed of a crash scene. There's not so much as a poof. The expensive, cherished spacecraft is just no longer there.

The fate of the Mars Climate Orbiter in particular proved not only heartbreaking but embarrassing. Any spacecraft can have bad luck with storms on Mars or mechanical failure. But an investigation showed that the spacecraft fell victim to a navigational mistake by an outside contractor in converting English units to metric. No one at JPL caught the error. It's hard to imagine that the JPL of the 1970s, which flew the fabled Viking and Voyager planetary probes, would have made such a blunder.

So now imagine that you are Bob Mitchell, project manager for Cassini, a $3 billion spacecraft, the cost shared with the European Space Agency. Cassini is a blast from the past, the last of the "Battlestar Galacticas," the huge robotic probes dreamed up in the 1970s and 1980s, before NASA adopted its famous and controversial "faster, better, cheaper" philosophy. Conceived in the early 1980s, Cassini is a dramatic reminder of how long it takes to design, build and fly a spaceship to the outer reaches of the solar system.

Mitchell couldn't sleep well for months as he waited for Cassini to arrive at Saturn on July 1. His team had to brake the spacecraft in a precise manner, enabling it to go into orbit around the planet.

Many things could have gone wrong. One of the biggest challenges was simply knowing where Saturn was, precisely. Sure, you can see it in the sky, but it's a billion miles away (and that's not just hyperbole). The distance makes it impossible to control Cassini in real time, because it takes more than an hour for a message to travel at the speed of light from Earth to the spaceship.

"It had to be just absolutely perfect," said Mitchell, who speaks with a quiet, square-jawed earnestness. "Any flaw had the potential, if not the high probability, of being mission-catastrophic."

You'd want to remember to convert your miles to kilometers in such a situation.

Complicating matters were those rings around Saturn. To take advantage of Saturn's gravity as an aid in braking the spacecraft, the engineers wanted to fly relatively close to Saturn's surface. But that would mean flying through the rings. They decided to aim for a gap between the F and G rings.

They had to thread the needle, shoot the moon -- or perhaps the new saying should be "shoot the rings." And they did just that, a perfect entry that allowed Cassini to become a Saturn satellite.

The views are nice out there. Cassini has taken fabulous images of Saturn's rings and the huge moon Titan. In approaching Saturn, Cassini managed to pass within 8,000 miles of an eccentric moon named Phoebe. It's a heavily cratered ball of ice that in all probability is an exotic thing known as a Kuiper Belt object, a planetoid of a type normally found on the fringe of the solar system, this one somehow captured by Saturn. Cassini's camera obtained an image of Phoebe of stunning clarity, offering planetary scientists an instant treasure that they will interpret for years to come.

In such a moment, humans are seeing something entirely new. This is the essence of exploration. Does it matter that astronauts weren't along for the ride? Maybe only in the degree of drama.

"It's different from walking on the moon," says Mitchell. "You just can't compete with that."

I, Human

Elachi, JPL's director, works in an office filled with a little boy's fantasy collection of model spaceships: Cassini, Magellan, Ulysses, the Spitzer telescope, the rovers, Odyssey, Mars Global Surveyor.

"This is good ol' Voyager, still flying," he says, pointing to a probe that visited the gas giants in the late 1970s and 1980s and is still beeping away as it heads toward interstellar space.

One might guess that Elachi, as head of JPL, would heavily favor robotic rather than human spaceflight. But he'd love to see humans operating a drilling rig on Mars, searching for microbes deep beneath the surface.

"There are certain things that robotics cannot do. We can make them more sophisticated, we can make them sing and dance, but you're still going to need human judgment in the end," he says.

Robots vs. humans is an old argument at NASA. The best feature of robotic craft is that you don't have to keep them alive. They aren't going to be fried by solar radiation during their interplanetary trek. They don't need oxygen and water and food. They travel light, compared with humans.

But advocates of human spaceflight argue that machines only do what they're told to do. No machine has anything like the dexterity of a human hand or the analytical power of a human brain. The speed of light creates an inevitable delay between Pasadena and Mars, so someone like Brian Cooper can't actually drive a rover in real time. It takes 12 minutes right now for a radio signal to make the one-way journey from our planet to the red one. As Elachi puts it, "You can't joystick things."

And there's one other thing a machine can't do. It can't feel anything. No joy, no wonder, no inspiration.

Going the Distance

There are lots of cosmic questions that NASA would like to answer. Are there oceans under the surface of Jupiter's icy moons? Are there Earthlike planets orbiting distant stars? Where might we find life in the universe?

The robotic program and a new generation of space telescopes will grapple with these mysteries. Space science has a bright future even if human spaceflight has a cloudy one. NASA recently launched a spacecraft toward Mercury. There are programs underway to send robotic probes to comets, to the moon, to Mars again, past Pluto to the outer edge of the solar system, and to the tantalizing moons of Jupiter. In roughly a decade, a planned telescope known as the Terrestrial Planet Finder may be able to obtain a direct image of an Earthlike planet in a different solar system.

The NASA road map for the future would have human and robotic spaceflight functioning in tandem. In some situations, rather than the machines supporting the humans, the humans would be supporting the machines, deploying and fixing telescopes and scientific instruments in space. It's not lost on NASA that one of its best moments in recent years came when shuttle astronauts managed to repair the Hubble Space Telescope.

NASA anticipates retiring the shuttle as soon as the space station is completed. Though the shuttle and station have powerful political allies, they are much derided among space buffs. In returning again and again to Low Earth Orbit -- just a couple of hundred miles above the surface -- NASA is hardly venturing into deep space. It's like a champion swimmer who only treads water.

"It's so clear. NASA's mission, what they do well, is exploration," says Louis Friedman, executive director of the Planetary Society, a space-advocacy group headquartered in a vintage bungalow a few miles from JPL. "That's why NASA's doing so well in the unmanned program -- they are exploring."

Perhaps the private sector will take over what has long been a government monopoly on the human kind of space exploration. There are entrepreneurs who are dying to rocket off the planet. But space is a harsh environment, alluring but deadly. It may be that biological intelligence will need to give way to artificial intelligence in the exploration of the universe.

Right now, all human beings remain on the Earth, save two: Gennady Padalka and Mike Fincke, who are circling the world about 240 miles above the surface, their names unknown to most of the 6 billion others.

The space program will surely find a way to survive. America would never want to cede the high frontier to some other country. The Chinese are actively pursuing an agenda that includes the possibility of Chinese astronauts on the moon.

There are many ways that the story of humans in space may play out.

"Mars theoretically could be modified," Elachi says. "A hundred years from now we could have the technology to put oxygen in the atmosphere."

A less fanciful idea would be to create an "interplanetary Internet," enabling people on Earth to connect to instruments and scientific stations on Mars and elsewhere. Elachi imagines classrooms doing experiments with such a system, but what might really drive the enterprise forward would be entertainment. Who knows, NASA merging with Nintendo, maybe?

What's certain is that, in just a few months, the scientists and engineers of the space program, both in America and in Europe, will do another of their amazing stunts. The Cassini spacecraft will drop a probe, named Huygens and designed by the European Space Agency, into the atmosphere of Saturn's moon Titan. Huygens will parachute to the surface and, via Cassini overhead, transmit images and data back to Earth.

Huygens may even survive the landing, either thudding on hard ground or splashing into an ocean of liquid methane or ethane. It might continue to operate for about a half-hour on that alien world before its batteries die and Cassini disappears over the horizon.

It's bold stuff. It's exploration. And if we can do something like that, then . . . maybe we can put a man on the moon.

An artist's rendering shows Cassini approaching Saturn. In July, NASA directed a perfect orbital insertion that made the craft a satellite of the planet. The Mars rover project has been a notable success for NASA since the Columbia disaster. Two travel the planet surface.The Endurance crater, as seen by the Mars rover Opportunity. Engineers can't control the rovers like a toy car; it takes 12 minutes for a radio signal to travel from Earth to Mars.Above, an artist's conception of Cassini shows the Huygens probe on its way toward the moon Titan. Below, a NASA diagram traces the path Cassini would take between rings to settle into orbit around Saturn. Cassini's close pass to the moon Phoebe allowed it to capture an image of stunning clarity, above. The spacecraft also sent back fabulous images of Saturn, below, after a seven-year trip through space. CHARLES ELACHI

Jet Propulsion Lab director

BOB MITCHELL

Cassini project manager