Early this year NASA had all but written off the Hubble Space Telescope, but today a robotic mission to replace worn-out batteries and gyros, and even to install new instruments, suddenly seems so doable that the agency is likely to ask for proposals to do the job in early June.
"I'm not saying it's a done deal," said Edward J. Weiler, NASA associate administrator for space science. "A lot of water needs to go under the bridge, but it's looking a lot better than it did two months ago."
NASA will have to decide within seven months whether to make the trip, because it needs three years to prepare a mission by the end of 2007, when Hubble's batteries are expected to give out: "If we haven't made a decision [this year], we'll lose the option," Weiler said.
In an interview at NASA headquarters, Weiler said the Goddard Space Flight Center received 26 responses to a Feb. 20 "request for information," inviting ideas for a robotic servicing mission. Goddard, in Greenbelt, is responsible for the telescope's engineering and maintenance.
The replies came from individuals, government agencies, specialty contractors and aerospace giants such as Lockheed Martin, and they offered schemes for everything from rendezvous and docking to software modification and robotic repair and servicing.
"There was no one silver bullet, no one right answer," Weiler said. "I was skeptical . . . but . . . the technologies we need are out there."
The proposals include the University of Maryland's "Ranger," a 25-foot stick-like robot that is tall enough to stand on a platform moored to Hubble and use its two mechanical arms to open up and reach every module and door on the telescope's bottom half.
The Canadian Space Agency's Special Purpose Dextrous Manipulator or "Dextre," has two 10-foot arms that pivot around a central "torso." The Canadians designed it to assist "Canadarm," the robotic arm in use aboard the international space station.
And NASA's own Johnson Space Center is developing "Robonaut," a humanoid robot with five-fingered hands and human-size arms whose eventual purpose is to replace astronauts during spacewalks.
Weiler said Goddard expects to decide by June 1 whether to formally request bids on a robotic mission. He said he did not know whether NASA would choose one contractor or several, run the project itself or seek a private-sector partner.
Bellow of outrage
"I suspect they're going to pick and choose to come up with a synthesis that's better than any of the individual plans," said aerospace engineer David L. Akin, who leads the Ranger project at the University of Maryland's Space Systems Laboratory. "It's the same with every mission -- the devil's in the details."
Five months ago, there were no details. NASA Administrator Sean O'Keefe on Jan. 16 canceled Hubble's fourth servicing mission, saying the shuttle could not fly to Hubble and still comply with new safety measures recommended after last year's Columbia disaster.
The decision provoked a national bellow of outrage from Hubble devotees. In March, O'Keefe asked the National Academy of Sciences to reassess servicing options, including a shuttle mission. The academy's findings are expected in late summer.
"I think the outpouring of public support helped NASA realize how important it is to keep Hubble's productivity at a high level," said Steven V.W. Beckwith, director of NASA's Space Telescope Science Institute, which sets Hubble's science agenda. Beckwith said he still hoped for a shuttle mission, but he acknowledged "great interest" in a robot that might be able to install new instruments.
Still, it will be risky. "You can't underestimate the complexity and the dangers," said former astronaut Jeffrey A. Hoffman, a Massachusetts Institute of Technology aerospace engineer who made three spacewalks to repair Hubble in 1993. "Suppose you open a door but can't put in the new instrument. Now you've got a light leak, and you've lost your telescope."
Hubble was launched in 1990, and is in orbit 360 miles above Earth. It operates on electric power drawn from batteries that are recharged by solar arrays. The telescope also has six onboard gyroscopes packaged in pairs inside basketball-size containers, and needs at least three of the six for position control. Engineers are working on ways to operate with two.
Deadline: End of 2007
NASA calculates that there is a 50 percent chance Hubble will have only one functioning gyro by mid-2006. At that point, controllers will still be able to hold the telescope steady, but scientific work must stop.
The deadline for servicing, however, is the end of 2007, NASA's best estimate of when Hubble's batteries will no longer be able to hold a charge. When that happens, Weiler said, the telescope will fail "within a matter of hours, or days."
Without a shuttle mission, NASA eventually must send a service module to dock with Hubble, if for no other reason than to guide the telescope out of orbit and to a safe crash landing, preferably in the open ocean.
If allowed to tumble out of the sky, Hubble's fiery remains could end up anywhere on Earth within 56 degrees of latitude straddling the equator, Weiler said, a huge band that embraces cities ranging from Miami and Hong Kong in the north to Rio de Janeiro and Brisbane, Australia, in the south.
"We don't want uncontrolled reentry, so we need an automated control and docking capability, which we don't have," Hoffman said. "We should have it. This will be money well spent."
NASA has $300 million set aside for de-orbiting. A servicing mission will cost more, but NASA does not yet know how much more. "We'll find the money," Weiler said. "It's not going to be three or four times as much," he added, because then "you'd have to ask, 'Why not build another telescope?' "
The module would probably dock with Hubble using the same mechanism that enabled the shuttle to lock to the bottom of the telescope on past servicing missions, Hoffman said. And once docked, there would be no reason for the module to leave, because sooner or later it would be needed for de-orbit.
Hoffman noted that the docking gear includes "plugs" that enable the shuttle to supply power to the telescope during servicing. The unmanned vehicle could use the same connections to mate the telescope to a new set of batteries that would remain in the module, he said.
A similar fix might work for the gyros, but Hoffman suggested a better way would be to install them inside the telescope when -- and if -- NASA replaces the Wide Field and Planetary Camera 2 with Wide Field Camera 3. The wedge-shaped module holding the camera can easily accommodate the new gyros, Hoffman said. And replacing the module only requires removal and replacement of two bolts and making some electrical connections.
This is where the robot comes in. Akin said he has been working on Ranger for 20 years, much of the time under NASA contract, "with a basic design philosophy to do Hubble-type tasks in the same way that astronauts do."
Ranger, controlled with a joystick from the ground, selects tools from a "tool belt" and plugs them into a fitting on the robot's hands. "I can't say we can do all the tasks today," he said. "But here's the hardware, and I know it can do the mission you want."
Canada's Dextre was scheduled to go to the space station in 2005, but the Columbia shuttle tragedy has delayed shipment, and Dextre is available. Canadian Space Agency science and technology adviser Jean-Claude Piedboeuf noted that all the robot's components -- "right down to the cables and lubricants" -- were designed to withstand the rigors of space.
NASA officials declined to discuss Robonaut because of the potential bidding competition, but their news releases suggest the robot, with human-like hands that can grip tools, could be ready in 2006 or 2007 -- in time for a Hubble mission.
Beckwith, at the Space Telescope Science Institute, said a robotic servicing mission would not be complete without installing the Wide Field Camera 3, capable of investigating the early formation of galaxies, and the Cosmic Origins Spectrograph, which will be able to study the distribution of matter between stars and galaxies.
"New batteries and gyros are fine, but that isn't what's made servicing such a success," Beckwith said. "The reason Hubble looms so large in people's consciousness is that we've improved it every time astronauts go up there."
Special correspondent Sumitra Rajagopalan contributed to this report.