Three years ago, a small robotic rover named Sojourner whirred across the rust-red Martian landscape and into the hearts of millions worldwide.
Now, scientists have a thumbs-up to aim another rover at the Red Planet. Call it Sojourner's big brother: Larger, heavier, and smarter than Sojourner, the new device will be a full-blown robo-geologist, complete with the high-tech equivalent of a geologist's hammer and magnifying glass.
It remains to be seen how endearing the coffee-table-size six-wheeler, scheduled for a 2003 launch, will become. But to Cornell University astronomer Steve Squyers, the rover represents "a huge step forward" in US efforts to explore Mars. Covering up to 100 meters a day for 90 Mars days, the rover would conduct the longest scientific trek ever attempted on another planet.
Its design testifies to the challenges researchers face when they want to explore a distant, alien world where human boot prints are unlikely to appear anytime soon. These range from the technical demands of building autonomous robots to mundane changes in personal habits, as scientists and controllers shift their work schedules from Earth time to Mars time.
Like the two Mars missions that failed last fall, the rover's task is to look for signs of water, says Dr. Squyers, the project's lead scientist. Water is fundamental to organic life. Evidence of its presence raises the possibility that life may once have gotten a microbial toehold on Mars - and perhaps lies buried beneath the surface today.
In June, researchers presented images from the Mars Global Surveyor that indicated water had once flowed on the planet's surface. Shortly after, other researchers presented geochemical evidence from a Martian meteorite that suggests Mars may have had larger reservoirs of water than previously believed.
The only way to get a better handle on the issue, Squyers says, is to go to the surface and poke around - via the new rover.
Where Sojourner sidled up to a rock and "sniffed" it with a spectrometer to glean its composition, the new rover will scratch and sniff: An abrasive tool will expose unweathered rock to a more sophisticated spectrometer.
A camera will gather images of crystal grains that would take a magnifying glass to see on Earth. And infrared and visible-light cameras will let researchers map the rover's surroundings and get some idea of their mineral composition.
Scientists' abilities to interpret this kind of data are coming along, according to Eric Baumgartner, who leads the mechanical and robotics technical group at the California Institute of Technology's Jet Propulsion Laboratory in Pasadena, where the rover is being refined for this and future missions.
Practice makes perfect
Practicing with a test rover named FIDO, the scientists had to analyze an unknown site they later learned was in central Nevada. "After about four or five days, they had the geology correct to about 80 percent," Mr. Baumgartner says. "It was pretty exciting because they found evidence of past water present in the rocks."
Yet FIDO was pampered by engineers at the test location - a luxury the 2003 rover will not have. Indeed, depending on where Earth and Mars are in their orbits, 11 minutes can elapse between the time a controller sends the rover a command and receives a "roger" back - enough time for it to topple over the edge of a canyon wall.
As a result, Squyers says, the craft is being designed to find its own way. "We want to say, 'Go 75 meters north northwest, then call us when you get there' and have the rover do that."
En route, the rover will stop every 20 centimeters to "look" at its surroundings with a pair of cameras and construct a 3-D map of the surrounding terrain. It should sense when it is in danger of overturning and adjust its course to avoid obstacles. Most important, Squyers says, "it must know when to stop."
The rover was one of two missions vying for the 2003 launch, according to Carl Pilcher, scientific director for NASA's solar-system exploration program. Agency officials announced the selection of the rover last Thursday.
The loser was an orbiter that would have been "a fabulous mission. I'd still like to see us fly it," he says. One reason NASA opted for the rover is that the next windows for a launch after 2003, in 2005 and 2007, would not be as favorable for trying to land instruments on the surface.
And while an orbiter might seem like a safer choice for an agency recovering from back-to-back Mars-mission failures, Dr. Pilcher holds that the rover presents a relatively low-risk approach. It will use the same kind of bounce-and-roll landing system that put the Mars Pathfinder/Sojourner mission safely on the surface. Unlike Pathfinder, which contained both a lander and rover, the 2003 mission's rover will land itself.
While Pilcher is reluctant to specify how lessons learned from the two recent failures are being applied to the 2003 mission, Cornell's Squyers says he sees improvements in several areas.
"We're clearly being given the resources to do the job right," he says. Sojourner cost $25 million, while the price tag for the 2003 rover is expected to be between $350 million and $400 million. In addition, he says, key elements - from instruments to the inflatable landing system - are flight-proven. Finally, the project has been subject to "very vigorous reviews," which will continue until the launch.
Indeed, NASA is even considering sending two rovers on the same rocket - each dropping to a different site on Mars. This approach, reminiscent of the Viking missions in the 1970s, has two advantages: The program gets twice the science for a relatively small added cost; and a second rover provides a back-up in case one fails.
NASA officials say they expect to decide on the issue within the next two weeks.
(c) Copyright 2000. The Christian Science Publishing Society