Dispatches, July 1997
Samad Hayati is looking through video eyes at what his six-wheeled robot sees as it picks its way through a martian landscape. "It's not red," says Hayati, who comanages the California-based Jet Propulsion Laboratory's Rocky 7 interplanetary rover project. "But it sure is desolate." The sand is flecked with black volcanic rock and pocked with shallow craters, one of which has been stumping Rocky 7, the JPL's star robotic rover. Each time Rocky tries to bypass it, he bumps on the rim and spins around, "hovering around the crater," Hayati laments.
Thankfully, this mishap didn't actually occur on Mars. It was a dry run, held in April on the parched bed of California's Lavic Lake. But even if Rocky's problems are not yet multibillion-dollar snafus, they do vividly animate the pitfalls of trying to rehearse for an arduous remote-control search for traces of life on Mars. Indeed, thanks to the recent frenzy over the possibility of such life, the rush to the red planet is currently underway. The first rover voyage, featuring a proto-Rocky called Sojourner, is due to land on the fourth of this month. Sojourner, however, will cover less than 100 yards. Finding signs of life on Mars requires a rover capable of traversing what Carol Stoker, who conducts field tests for NASA's Ames Research Center, calls "features of interest thousands of kilometers in size." Thus the pressure is on scientists not only to develop more advanced rovers, but to ensure that they can handle all manner of "Mars analogs," or earthly approximations of the stark martian landscape. And right now, it seems they have their work cut out for them.
Of the three outfits doing the bulk of the research, only the JPL has yet to take its analog tests far afield. Such trips, Hayati says, would be "superfluous," given that Rocky 7 still can't tell if the occasional creosote bush is a rock to avoid or a clear path to cross. Carnegie Mellon University, another NASA contractor, has been more bold in its testing, taking its spiderlike Dante robots to climb where humans dare not — inside active volcanoes. But the first Dante climbed just 27 feet down Antarctica's Mount Erebus before its tether broke; Dante II made it to the bottom of Alaska's Mount Spurr, but then slipped on the way back and ultimately lost seven of its eight legs.
This put an end to walking-rover research, but wheeled models are faring better. Using a stout, Russian-built Marsokhod rover, Ames researchers have conducted successful tests across Hawaii's Kilauea caldera and the harsh Arizona desert. Ames has also used Marsokhod's underwater counterpart to refine its "telepresence" navigational systems, diving beneath the Antarctic ice sheet to prove, as researcher Chris McKay puts it, that the rover could "do science in remote locations" via satellite control. Still, Marsokhod is not without problems of its own: At 330 pounds, it may be too heavy to be affordably shuttled over an expanse of more than 34 million miles.
Nonetheless, the telepresence technology used to guide Marsokhod has given rover researchers hope. If they can somehow meld it with the maneuverability of Rocky 7, we may well see a robot cruising Scarsdale-to-Soho distances across Mars. "These are only preliminary tests," says Hayati, as Rocky 7 struggles to overcome its crater hang-up. "We've still got a few years to work the kinks out."