Details Emerge in Robotic Plan to Service Hubble

BALTIMORE - An astronomer-astronaut who hasjourneyed twice to work on the Hubble Space Telescope said Monday that NASAAdministrator Sean O'Keefe is conceptually sold on the idea of a roboticservicing mission.

Former astronaut John Grunsfeld, now NASA's chief scientist, also provided newdetails for how the possible reprieve came about. He discussed the pros andcons of robotic versus astronaut servicing and ticked off a priority list forany robotic effort.

Hubble has just two or three years of observing left in itsbatteries and pointing gyroscopes. A decision on a possible mission is expectedby early June.

While astronomers and the public have spent the past threemonths bemoaning O'Keefe's unilateral decision not to send a space shuttle backto repair and upgrade the orbiting observatory, Grunsfeldwas contemplating alternatives. He has personally grappled with installation ofnew Hubble equipment during daring space walks.

O'Keefe told lawmakerson April 21 that the robotic mission looked promising based on 26 responses toa call for ideas.

A robot can definitely do the work, Grunsfeld said. But ifthe telescope is to be saved, a final commitment must be made soon to allowtime to plan a mission unlike any ever undertaken.

Lobbying the chief

NASA announced Jan. 16 that a crewed mission to fix Hubbleand add new and powerful instruments would be cancelled due to safety concernspresented in the wake of the Space Shuttle Columbia disaster.

"The decision hit me in the head like a two-by-four,"Grunsfeld told astronomers gathered here at the Space Telescope ScienceInstitute (STScI) to ponder how to make the best use of Hubble's final years.After a couple weeks of depression over O'Keefe's initial decision, Grunsfeldwas commiserating with a NASA engineer and learned that robotic servicing mightbe practical.

So he took the idea straight to the top.

Grunsfeld and Ed Weiler, NASAassociate administrator for space science, put the suggestion to O'Keefe. Theyargued that the technology needed to carry out space-based robotic repair fitneatly with the requirements of President Bush's new vision of developingrobotics and other capabilities necessary for setting up a Moon base andsending astronauts to Mars. That means it would fit logically within the spaceagency's budget, which Bush wants restructured to support the new long-termgoals.

"Mr. O'Keefe was totally sold," Grunsfeld said."This took about five minutes."

Already, Grunsfeld said, equipment used by astronauts totrain for past Hubble servicing missions has been shipped to NASA's GoddardSpace Flight Center and two other locations where space-based robotics wereunder development for other purposes.

No decision has been made on how ambitious the mission mightbe, should O'Keefe approve it.

Meanwhile, astronomers had just been coming to grips withthe expected loss of Hubble. Several of them reiterated at the meeting, afour-day discussion about which observing projects Hubble should undertake inan abbreviated life, why Hubble is so vital to investigating star and galaxyformation, exploring extrasolar planets and peeking at the earliest epochs ofthe cosmos.

No observatory seriously planned for launch anytime in thenext 10 years can replace the optical and ultraviolet capabilities of Hubble,they said.

"The technology seems to be farther along than I hadrealized when the idea first came up" for a robotic fixit venture, saidDavid Leckrone of Goddard.

Robots vs. astronauts

"I don't care how we service Hubble" as long asscience continues to flow from it, said Grunsfeld, who is trained as anastrophysicist.

It has not yet been determined whether there is enough timeto plan for having robots successfully install new instruments that are alreadybuilt and were intended to make Hubble more sensitive and useful than ever, orif the mission would instead be limited to keeping Hubble functioning at itspresent ability.

But Grunsfeld said employees at various NASA field centers"are just supercharged" to try and make it all happen.

In an interview, Grunsfeld said there's a clear prioritylist. The first would be to attach a device of some sort that would ultimatelybe used to de-orbit Hubble into the ocean. That part of the mission wouldfulfill a requirement that had already been in place to safely bring Hubbledown sometime in the next decade or so.

The second priority: "Don't break the Hubble," hesaid, on the assumption that it is still operating when the robot arrives.

Third would be to replace the batteries, which are likely togo before the gyroscopes, according to the latest analysis. Replacing thegyroscopes is a close forth on the list.

New batteries and gyroscopes would buy about six years ofservice from the installation date.

There is a mechanical issue that could work to the favor ofastronomers: Gyroscopes might be attached to the outside of the observatory,but that would be less than ideal due to problems of stability. A more stableoption would be to mount them inside one of the new instruments and install thewhole setup, achieving longer life andmuch better science, Grunsfeld said.

There are pros and cons to using astronauts versus robots.From experience, Grunsfeld said equipment sometimes gets stuck while beingswapped out. Astronauts can feel what's going on, stop, and make adjustments.On the other hand, "robots can do really pure motions." And a robotcan remember the exact movements needed to remove a part, thenduplicate the motion in reverse to install a replacement.

The process would not be automated. Instead, servicingHubble would be a bit like an orbital video game.

Grunsfeld explained that the robot would be controlled fromthe ground, in real time, by someone familiar with the telescope -- perhaps aformer Hubble-servicing astronaut like himself. He added that even if a roboticmission did not fully succeed, engineers would learn plenty to apply towardfuture efforts at remote operations on the Moon and Mars.

"So it's a win-win situation," he said.

Fresh optimism

Astronomers who have felt left out of the decision to cancelthe manned servicing mission were delighted to hear the upbeat report on thepossible stay of execution for Hubble.

"NASA clearly feels the need to do something,"said Steven Beckwith, director of the STScI, which operates Hubble for NASA.Beckwith and his staff were shocked when NASA decided not to send astronautsback to Hubble. He has since been pleased with "an outpouring of publicsupport" for the telescope that he called unprecedented in the world ofscience.

This week's meeting, the 18th such springgathering at the STScI, was planned prior to O'Keefe's January announcement.Its title, Essential Science in Hubble'sFinal Years, turned out to be far more prescient than its plannersexpected.

Grunsfeld was not on the speaker's list handed out toconference attendees. The astronaut said O'Keefe had wanted to come, "tolook all of you in the eyes" and explain the earlier decision not toreturn to Hubble. But the NASA chief was unavailable, having been called tospeak before a Bush commission designed to set a course for meeting the newhuman spaceflight goals.

Astronomers have been concerned that O'Keefe's decision wasnot just about safety but was made to help shift the agency's course fromscience to human exploration. Not for the first time, Grunsfeld said that's nottrue.

"I got a very clear statement [from O'Keefe] that itwas not about the budget," Grunsfeld said.

The decision did reflect the requirements of the CAIBcommission that reviewed the Columbiadisaster and set guidelines for NASA's return to flight. It also included anextra measure of safety, based partly on what Grunsfeld called O'Keefe'sintuition that NASA should have a second shuttle ready to fly a rescue missionin the event the crew of the servicing mission found itself in a faultyshuttle.

Grunsfeld said time was of the essence in the originaldecision, too. Even if the first shuttle flight occurs next spring, astentatively planned now, it would be unrealistic to expect that after a fewtest flights -- in which problems might be discovered causing further delays --that Hubble could be serviced before its batteries or gyros fail. He said ahuman journey to Hubble would be at least fifth on the return-to-flightpriority list, behind shakeout flights and at least one trip to theInternational Space Station.

Best bet?

Grunsfeld said a robotic mission is, after considering allthe factors, the more likely to be pulled off in time. "If we need to dosomething, we need to do it fast," he said.

Beckwith, the STScI director, cautiously agreed that arobotic mission might turn out to be the best bet. But he said he press formore than just the installation of batteries and gyroscopes. One of the reasonsHubble is such a great observatory, he said, is that previous missions haveadded new capabilities. One of the two planned new instruments for Hubble wouldmake it 10 times more capable at infrared observations, other researchers havesaid.

"We don't know yet what robotic servicing means,"Beckwith told his staff. "We should be optimistic."

Join our Space Forums to keep talking space on the latest missions, night sky and more! And if you have a news tip, correction or comment, let us know at:

Robert Roy Britt
Chief Content Officer, Purch

Rob has been producing internet content since the mid-1990s. He was a writer, editor and Director of Site Operations at starting in 1999. He served as Managing Editor of LiveScience since its launch in 2004. He then oversaw news operations for the's then-parent company TechMediaNetwork's growing suite of technology, science and business news sites. Prior to joining the company, Rob was an editor at The Star-Ledger in New Jersey. He has a journalism degree from Humboldt State University in California, is an author and also writes for Medium.