Is a Robotic Servicing Mission the Answer for Hubble?
By Walter Cunningham
Two days after President George W. Bush announced his "Moon, Mars and beyond" initiative, NASA disclosed it was canceling the last Space Shuttle mission to repair and upgrade the Hubble Space Telescope. Since then, the public outcry over the cancellation has drowned out support for the new initiative.
In response, NASA's managers have proposed a robotic mission to extend Hubble's life and, in the process, demonstrate new technologies that can advance its broader space exploration agenda. In reality, they are risking science, exploration, taxpayer dollars, and possibly even human lives to avoid flying the Space Shuttle one more time.
Ever since a human servicing mission fixed its initial blurry vision, the Hubble Space Telescope has been one of the most remarkable facilities in the history of science and, with the now-cancelled equipment upgrades, would continue its breakthrough scientific observations into the next decade. NASA says the Webb Telescope will replace the Hubble in 2011. The Webb should be a wonderful instrument if its highly-experimental design works, but its infrared capability would be complementary to the Hubble's visual and ultraviolet range, not a replacement for it. At this stage, the Webb is still just a design, not a proven observatory, and it will be doing well to reach orbit by 2015.
The Hubble was designed to be serviced by Shuttle astronauts, not by robots. Prior to the Columbia tragedy, the Shuttle flight manifest included a planned Hubble servicing mission (SM4) in 2004 or 2005 and another Shuttle mission in 2011 to retrieve the telescope or attach a propulsion stage that would enable a controlled deorbit.
Following Columbia, SM4 was delayed until 2006 and the 2011 Hubble retrieval/deorbit flight replaced by a robotic deorbit vehicle. This was predicated on the astronauts on the SM4 mission attaching devices to Hubble that would make the subsequent robotic end-of-life mission easier to accomplish. This allowed NASA time to develop and demonstrate currently-unproven robotic rendezvous and servicing techniques before they were needed in 2011.
In January, citing safety concerns for Shuttle astronauts, NASA cancelled SM4 and moved up the Hubble robotic mission to 2007. This "faster, smarter, sooner" robot will now also be required to perform the servicing tasks originally scheduled for SM4. That is, doing twice as much with half the time to prepare, and a predictable impact on cost.
NASA officials are asserting it is "possible" a robot will work. Of course, it is also possible the robot will fail and may cost more than a Shuttle flight.
According to former astronaut John Grunsfeld, now NASA's Chief Scientist, the priorities for a robotic mission are:
1. Attach a deorbit motor that would eventually be used to deorbit Hubble into the ocean;
2. "Don't break the Hubble," assuming it is still operating when the robot arrives;
3. Replace the batteries; and
4. Replace the gyroscopes (a close fourth).
If successful, this plan could keep Hubble alive a few more years, but there is no credible plan to upgrade its scientific capability, which was an integral part of SM4. Based on the historical database there is a high probability that even this less-challenging robotic servicing mission will fail. So far, there has been little public discussion of potential unintended consequences of such an outcome.
Robots are Unproven and Risky
Automated/robotic rendezvous and docking is an absolute requirement for a robotic Hubble servicing mission. The Russians have done this for 25 years, but NASA has no spacecraft capable of performing this mission, and it is debatable if we have the time or the funding to develop one. The history of robotic space technology suggests that a rushed robotic servicing mission, using unproven technology, is expensive and likely to fail. Past efforts at automated operations include:
NASA's Orbital Maneuvering Vehicle was cancelled in 1989, when the estimated cost exceeded $1.3 Billion. Europe's Automated Transfer Vehicle, originally scheduled for completion in 2002, is three years late and will cost over $1 Billion.
Japan's Hope Transfer Vehicle is running six years behind its original completion date.
NASA's program to demonstrate automated rendezvous, for two spacecraft explicitly designed to rendezvous and dock with each other, is running six months late.
A number of NASA's past attempts at a robotic or partially-automated approach have resulted in satellites being incapacitated or left in a useless orbits. Intelsat VI, Syncom IV, Palapa-B, and Solar Max satellites would all have been failures if direct, "hands-on" intervention of humans in space had not turned them into successes.
Safety and the Risk to Human Life on earth
The risk to astronauts was the focus of the Columbia Accident Investigation Board and has been blamed for the cancellation of SM4. A "safety" issue that has not been discussed in the public debate over Hubble's future is the increased risks that 3rd-party innocent bystanders on earth may be exposed to if NASA persists in a robotic servicing mission.
Absent a controlled reentry of the telescope, there would be a 1 in 700 risk of casualty to persons on the ground. NASA's historical standard has been 1 in 10,000. Since a rushed and complex robotic servicing mission is less likely to successfully install a deorbit motor on the Hubble, it is more likely to result in an uncontrolled re-entry, thereby increasing the risk of "loss of life" on the ground. Whether a robotic servicing mission would be reliable enough to meet the accepted standard, or whether NASA is still committed to meeting this standard, is unknown.
The Hubble was not designed to be robotically docked with a robot, let alone to be serviced robotically. And since robotic on-orbit servicing has not been demonstrated, any such servicing mission is clearly a risky program. Therefore, NASA must have a backup plan for a failed Hubble robotic servicing mission. A failed robotic mission, without an announced backup plan, could result in a repeat of the 1979 public fear over Skylab's death plunge. This would likely lead to a public outcry for a Shuttle servicing mission to Hubble. At that point, public policy makers would have a Hobson's choice among three painful alternatives:
1. Attempt another robotic de-orbit mission, using improved or altogether different hardware and software than the first failed attempt. Whether this could be done before Hubble begins to tumble, making a robotic approach impossible, is unclear.
2. Ignore the increased chance of a ground fatality and allow Hubble to make an uncontrolled re-entry into the Earth's atmosphere. The estimated 1 in 700 chance of a fatality would not only risk life, but expose the government to potential claims for property damage from whatever nation debris may fall on.
3. Conduct a rushed Shuttle flight, with the same risks as SM4, but which would produce only a safe de-orbit and none of the scientific benefits of SM4.
The unintended, but predictable result of attempting to reduce risk to human life by robotically servicing Hubble may be a further risk to human life, either on the ground or in space, or both, and a greater cost to taxpayers than the manned servicing mission.
NASA's January 2004 cancellation of SM4 has these negative impacts on the Hubble end-of-life Mission:
The challenging robotic de-orbit mission originally planned for 2011-12, will have to be carried out 3-4 years earlier, because once Hubble's batteries or gyroscopes fail, the telescope will tumble and be much harder, if not impossible, to robotically capture.
Because SM4 will not have prepared Hubble for a robotic rendezvous and capture (e.g., installation of optical targets and other aids), the robotic mission will be more difficult to accomplish even if it was not moved up on the schedule.
The Cosmic Origins Spectrograph and the Wide field Camera 3, already built at a cost of nearly $200 million are effectively dead, since they would have to be placed deep inside the telescope after the removal of existing instruments. This is considerably more difficult and risky than simply bolting new equipment to the outside of the observatory and connecting a few cables, as could be done for the gyros and batteries. (This would replicate most of the originally planned SM4 upgrades.)
NASA says their first priority must be "return to flight," and they should get that behind them before addressing Hubble. But if they wait until the Shuttle is flying before addressing the SM4 mission, it will all be academic. At that point, Hubble's batteries and gyros will almost certainly fail before a manned servicing mission could be executed. This will certainly be true if the agency is planning three test flights before returning to operational missions, as some have stated.
It is even possible that, with all the odds against its success, the Hubble robotic servicing mission will eventually be cancelled. It makes you wonder if it is intended to save Hubble or to save NASA management from public criticism.