Originally created 05/03/04

Robots to the rescue



COLLEGE PARK, Md. -- Think Edward Scissorhands, but with bolt drivers and pliers for hands and a giraffelike neck topped with a pair of cameras resembling black cratered eyes instead of the handsome head of Johnny Depp.

Could this strange robot take the place of astronauts in fixing the Hubble Space Telescope? NASA is yearning to find out.

With astronauts banned from Hubble because of space shuttle safety concerns, the University of Maryland's Ranger robot could conceivably save the day by installing fresh batteries and other life-sustaining parts on the observatory.

Or if not Ranger, then Robonaut, NASA's very own humanoid robot, or the Canadian Space Agency's Dextre, a two-armed robot intended for the international space station, or any number of other robots under development that could blast off aboard an unmanned rocket in three or four years.

While astronomers and astronauts may wince and scoff at the prospect of a machine working on their beloved Hubble, the robot crowd can barely hide its glee over NASA's search for a mechanical deliverer.

First stop: Hubble.

Next stop: moon, Mars and beyond.

The technology is here and the time is now, says David Akin, director of the University of Maryland's space systems laboratory and leader of the team that created Ranger. He estimates the technology exists to do 90 percent to 95 percent of whatever NASA wants at Hubble or the space station - or on the moon, its new target destination.

"If NASA waits until robots become servants in your house, they're way far behind the power curve," Akin says. Both Hubble and the space station will be long gone, "and it's going to be 2050 and we're still going to be talking about how nice it would be to go back to the moon."

NASA's associate administrator for space science, Ed Weiler, is becoming more and more a believer that a robot could extend Hubble's life. He's even considering making the next-generation James Webb Space Telescope modular so robots could replace parts; the telescope will be launched in 2011 to a point 1 million miles from Earth, well beyond human reach.

"I have new respect for robots, especially after the miracle of landing two robots on Mars and actually fixing one of them 100 million miles away," says Weiler. "So I've got to believe robots have a role, just as I believe humans have a role."

Akin prefers a pair-up, too. His research has found that together, a spacewalking astronaut and robot would be much more efficient than individually.

Ranger, in fact, was conceived as an astronaut's assistant, to serve as a third and fourth hand for Hubble spacewalkers and thereby reduce the time - and inherent risk - for humans outside. But with astronauts out of the space telescope picture, Ranger ought to take a solo swing at it, Akin says.

NASA estimates Hubble will likely stop observing the cosmos by 2007 or 2008 unless someone or something gets there before the batteries die.

For Akin, more than Hubble is at stake.

"I would like to think somebody at NASA realizes that to do humans on the moon and Mars, you're going to need robotics to set up lunar bases, to build transfer vehicles. To relieve the crew of having to do the grunt work of toting and carrying and so forth, you need dexterous robotics," he says.

"Everybody's willing on kind of a high-level conceptual basis to say, 'Yeah, that's absolutely true.' "

But while NASA has commissioned all sorts of computer graphics showing astronauts and robots working together, Akin notes, "they haven't been willing to put a penny into actually making it come true."

That may be about to change because of Columbia's demise and the countdown to Hubble's doom.

The Feb. 1, 2003, catastrophe is forcing NASA to find ways for shuttle astronauts to patch potential holes in their orbiting ships once flights resume next year. NASA's backup plan, failing successful repairs, is for a shuttle crew to seek refuge at the space station.

Astronauts at Hubble could not get to the station if something went wrong, and so NASA Administrator Sean O'Keefe in January killed the last telescope servicing mission on the shuttle books and decreed that the spaceships fly solely to the station.

Taken aback by the public outcry over abandoning Hubble, NASA put out the call for robots in March and 26 ideas were submitted. The space agency also sought the advice of the National Academy of Sciences, at the urging of members of Congress.

In mid-April, the academy established a committee to assess all the options for extending Hubble's life, including a shuttle mission. Among the 20 committee members: three former astronauts, including a spacewalker who worked on Hubble in 1997 and a pilot who helped deliver it to orbit in 1990; two Nobel physicists; a retired shuttle and station program manager; a Columbia accident investigator; and an artificial intelligence expert.

The committee will convene in late May and, hopefully, issue recommendations by late August - no later than November, says Joseph Alexander, director of the academy's space studies board. Typically, an academy study takes a year or more, but there's no time to waste in this case, he says.

NASA wants to settle on a course of action before fall.

"The worst possible scenario, if you really want to be a negative thinker, is that we fight and fight with various entities for the next year on shuttle versus robots, we make a decision to do the robot and because we started late, it's scheduled in 2008 and then the batteries fail in 2007," Weiler says.

Carnegie Mellon University's Red Whittaker, whose robots have cleaned nuclear power stations, worked as museum tour guides, trekked the Chilean desert and hunted for meteorites in Antarctica, considers a Hubble servicing mission just as feasible.

"One of the wonderful things about this era in robotics is that a lot of these great ideas are going from the laboratory and into the world, or in this case, to worlds beyond," Whittaker says.

Skyworker - conceived at Carnegie Mellon as an autonomous inspection, assembly and maintenance robot for high-orbiting solar power plants - is nowhere near ready for space prime time. Yet the thought of speeding development to help Hubble entices Whittaker.

Flying to Hubble's rescue also captivates the Robonaut team at Johnson Space Center in Houston.

Robonaut, supported in part by the Defense Advanced Research Projects Agency for possible battlefield use, resembles a spacesuited human from the waist up and can even do some tasks on its own with its five-fingered hands - like identifying, picking up and using a wrench. It can also plug in a phone jack and has shaken congressmen's hands.

But a space-qualified unit would need to be built, fast. The two existing Robonauts - one permanently affixed to a lab pedestal, the other periodically touring the country atop a Segway - are intended for ground use.

While Ranger lacks Robonaut's humanlike hands - a tool simply plugs into each wrist socket - it is further along in terms of flight readiness.

And even closer to completion is Canada's Dextre. It's supposed to be launched to the space station around 2007 to supplement the Canadian robotic arm already there, providing greater dexterity for outside maintenance work.

But Dextre would need to be revamped to tackle Hubble, the Canadian Space Agency says.

Seventy percent of a space-certified Ranger already exists, albeit in pieces, at the University of Maryland, just a few miles from Goddard Space Flight Center, which is reviewing Hubble's robotic options. After NASA pulled its funding in 2002 in a round of budget cuts, the robot was dismantled and stored in cabinets.

A test replica, though, occupies center stage at Akin's space systems lab. The 8-foot aluminum, stainless-steel and titanium robot - neckless and headless since gravity would cause the 4 1/2 -foot extension to shake - is anchored on a simulated shuttle cargo bay carrier. It comes to life, with a nudge of a joy stick.

Before choosing the best robot, if any, for the job, NASA must decide what to attempt. The canceled servicing mission by astronauts was ambitious and "even Robonaut would struggle with elements," says Chris Culbert, deputy chief of Johnson's automation, robotics and simulation division.

Besides popping in new batteries, the robotic repairman probably would be directed by ground controllers to install motion-control gyroscopes. It would be easier to slap new ones on Hubble's exterior, rather than open the doors and replace the old ones.

Depending on how aggressive NASA wants to be, the robot also could try to install a $176 million pair of cameras. But one camera would require opening one of Hubble's dreaded doors, which might be warped and cantankerous from their years in space.

"If you supply power and you supply gyros, wouldn't it be idiotic if you opened the doors, put in a new instrument and can't close the doors and you've got scattered light all over the place?" Weiler says.

Hubble spacewalker Jeffrey Hoffman had a tough time shutting a misaligned telescope door a decade ago. He believes "something with the sophistication of Robonaut" could succeed as long as the doors are not warped and behave properly.

Now a Massachusetts Institute of Technology professor, Hoffman just hopes NASA does not underestimate the risk to the mission since the robotic systems have never flown in space.

Weiler figures a bare-bones robotic mission would cost at least $300 million; the more attempted, the higher the tab.

Doing nothing, though, is not an option.

At the very least, Weiler says, NASA needs to launch and dock an engine and propulsion module to steer the 25,000-pound telescope into a Pacific grave a decade from now.

"The Number 1 priority is find a way to make sure that Hubble doesn't land in Miami or Calcutta or Mexico City," says Weiler.

Weiler expects holdouts for a manned servicing mission, no matter what.

"The beauty of this robotic option, if we can show it's viable, is it takes the safety issue off the plate," he says. "I'm sure there will be debates until the end of the universe. But that's the point. The batteries won't last that long."

On the Net:

Space Telescope Science Institute: http://www.stsci.edu/institute/

University of Maryland's Ranger: http://robotics.ssl.umd.edu/ranger/index.shtml

NASA's Robonaut: http://vesuvius.jsc.nasa.gov/er-er/html/robonaut/robonaut.html