NASA's Martian dowsers

Call NASA's Mars exploration efforts "all wet," and you would be paying the agency's scientists a big compliment. More than any other single substance, water reveals fundamental details about the planet's climate, its ability to sustain life, and its potential resources for future manned missions, according to Dr. Dan McCleese, chief scientist for the Mars Program at NASA's Jet Propulsion Laboratory (JPL). "Water is critical in each category," he says. "It drives our entire understanding of the planet." That understanding crested last June when images from an orbital camera aboard the Mars Global Surveyor suggested the recent presence of liquid water on the planet's surface. Yet far from quenching NASA's thirst, these images provided just a taste of what's to come. "We're learning today from the Global Surveyor that we're still very much in the reconnaissance phase," McCleese says.

For its next Mars mission, slated for 2001, NASA will continue to search for signs of water remotely-using an updated Global Surveyor orbiter fitted with more sophisticated imaging systems. Ultimately, however, the agency's high-tech water witching will require a combination of remote sensing and technologies that will go after geological and chemical samples on the ground. "To date, we have good, sometimes very good, remote data," notes Scott Hubbard, director of NASA's Mars Program. "What we still need is the 'ground truth,' which requires us to get next to a rock and make measurements."

Last year's mission misadventures forced agency officials to reconsider their Mars strategy, which centered on technologies to collect and return scientific samples. Hubbard explains that the failure of the Polar Lander mission caused the agency to temporarily limit its "appetite" for all but orbital missions-which are less risky and less costly than landed missions. Indeed, the 2001 mission initially included plans for a lander, not just the orbiter in its current incarnation.

With the 2003 mission, however, NASA plans to go back to the surface with two new rovers that will land directly on the surface inside a Pathfinder-influenced landing system-which employs a parachute and airbags for a bounce-and-roll landing that's not so much gentle as manageable. Weighing in at nearly 150 kg, with a range of up to 100 meters per Martian day, these rovers or "mobile landers" will outweigh and outpace the small Sojourner rover that arrived on Mars as part of the 1997 Pathfinder mission. They will also play a larger scientific role than the Sojourner, thanks to an on-board suite of instruments for analyzing rock and soil samples (see sidebar). "This mission will give us the first ever robot field geologist on Mars," says Hubbard.

Looking beyond 2003, NASA's "follow the water" strategy still requires a massive technology development effort. Says Hubbard, "The technology landscape really opens up after '03."

The right samples. Of the technical challenges faced on Mars, one of the toughest relates to the seemingly workaday task of gathering and returning samples. "It's an extraordinary technical challenge," McCleese insists as he lists attributes that future rovers will need: To get to the sample, the next rover will need on-board navigation and hazard-avoidance systems. To pick the "right" samples, it will need some built-in science smarts-"something more than 'I know a fossil when I see it,'" he says.

Finding the "right stuff" represents only half the battle. Rovers must also possess the means to harvest samples on or below the Martian surface. And returning the samples won't be easy either. McCleese points to all the interconnected steps required to store the sample, launch it off the surface with a small lander-based rocket, rendezvous with a waiting spacecraft, and lastly return it to Earth safely. "Each of these steps is a mission in itself," he says.

Meeting all these goals will call for rovers to evolve substantially. They will have to become autonomous "science platforms," McCleese believes. These future rovers will gain a measure of autonomy via advances in computing power, decision-making software, and telecommunications. "The notion of an autonomous rover is really very new for us," he says.

They will have to carry automatic sample-harvesting equipment, and NASA has in fact developed a keen interest in drilling technologies. Miniature drills from Honeybee Robotics (New York) have already been developed for coring rocks, but NASA also plans to go deep into the subsurface. In the short term, the agency's scientists want to drill in the 10-to-200 meter range-still shallow enough for some sort of robotic approach, McCleese reports. "We also believe liquid water may be five kilometers below the surface." Getting that deep may require human handiwork-space roughnecks-so it won't happen anytime soon. But NASA has gone so far as to consult the oil industry and has hired an oil-drilling expert from Chevron. "The oil industry is teaching us about measurements we need before we even attempt to drill," says McCleese.

As for payload, the descendants of today's rovers will likely sport Mars-ready scientific instruments to enable in-situ chemical or physical analysis. "We want to bring the capabilities of lab instruments to Mars," says McCleese. To take one example, NASA wants to develop instruments that can date samples on Mars. "We can do a good job at relative age, a poor job at absolute age," says McCleese. Looking further out, he envisions instruments that can detect bio-signatures through chemical or physical analyses of organic substances. "These instruments haven't been developed yet except in heavy and expensive lab versions."

While not all of the technologies McCleese envisions have been developed yet, a harbinger of what's to come does exist in FIDO. Short for Field Integrated Design and Operations, this robotic vehicle was initially developed to test concepts for the 2003 and 2005 sample return missions. Developed for multi-kilometer travel, this six-wheeled robotic vehicle weighs about 60 kg and measures 100x75x45 cm. It bristles with instruments-including a bore-sighted near-IR spectrometer, a microscope mounted to a robotic sampling arm, Raman and Moessbauer spectrometers, and a multi-spectral stereo mast camera. It also features an automated mini-corer and sample caching system. Late last spring, the FIDO rover took to the field-in the Nevada desert-for its first trials.

A soft landing. NASA's sample-analysis-and-return goals also trigger the need for new landing technologies. As rovers evolve into McCleese's "science platform," they will overflow with analytical instruments and heavy sample-return equipment. And getting rovers within striking distance of their targets will require smarter landing systems. "One of our major challenges is to develop robust, adaptive landing technology," Hubbard says.

To cushion the increasingly weighty payloads, NASA will likely have to come up with a replacement for the airbags it has used since the Pathfinder mission. "Airbags don't scale well with delivered mass," McCleese says, adding that they've nearly reached their practical limit with today's loads of 600 to 1000 kg. "We're already to the point where airbags become cumbersome and their effectiveness diminishes," he says. Likely airbag alternatives include revamping of the legged systems employed in lunar landings and also a "Frisbee-shaped" pallet landing system that can accommodate uneven or sloped terrain. Both of these landing concepts are currently under investigation at JPL, McCleese reports.

The next generation of landing craft will also require a better sense of aim. Hubbard notes that current landing technology can get within 10 to 100 km of a target, but NASA plans call for accuracy in the hundreds-of-meters scale. And he adds that this tighter landing accuracy "requires tools we don't have right now." NASA's list of possible solutions in this department includes adaptations of surface-mapping technology developed by the Department of Defense. Hubbard also predicts that faster on-board computer and neural net software will aid future landers as they avoid hazards and make real-time navigation decisions.

The human connection. While the data collected in robotic missions may be an end in itself, robotic missions also represent giant steps toward human voyages. "We have an integrated strategy," says Hubbard, explaining that today's early thoughts about human exploration already build on lessons learned from robotic missions. "Robotic missions could pave the way for human missions," agrees McCleese.

And when he says "pave the way," he means it-literally. He predicts that the future robotics will focus not just on science but on practical tasks such as building the habitat and other infrastructure components needed for human settlement. Thanks to robotics, he says, "the first humans on Mars can be explorers, not civil engineers."