Hubble Robotic Servicing and Deorbit Mission (HRSDM)

The loss of the Space Shuttle Columbia in February 2003 caused a reexamination of the capabilities of the Space Shuttle.  As a result of this, the decision was made in January 2004 to not service HST any longer with a human crew, but to allow it to decay gradually out of a operational state.  After much behind-the-scenes work, in October of 2004, NASA changed directions slightly, and we formed a team consisting of MDRobotics, Lockheed Martin and the Hubble Project to build a robotic vehicle to service the Hubble Space Telescope.

Announcement in (10/05/04)

Announcement in comspacewatch (10/06/04)

Announcement on MDR's website (10/06/04)

This major event was preceded by several months of effort on the part of our team to convince NASA that this mission could be done with available technology.  Prior to the loss of the Columbia, it was already the plan to launch an unmanned spacecraft to deorbit HST into the Pacific Ocean to prevent an uncontrolled reentry of Hubble.  This is because Hubble is so big and heavy that it will not burn up completely in the atmosphere.  It is possible that a part would fall out of space and hit land.  As a result, a Shuttle or a rocket vehicle was always planned for retrieval of Hubble.  It seemed feasible to us to include a robot on this mission along with new replacement hardware so that the telescope's life could be extended and it could be safely deorbited at the end of the mission.

Concept graphic of the HRSDM.  A new vehicle would fly up with a robotic system to grapple and
service HST.  The brown part is called the Ejection Module (EM), and contains the new instruments,
and the robotic system.  This part would be ejected and only the blue part would remain.  This latter
Deorbit Module (DM) will have a rocket motor to push Hubble out of Earth's orbit when it is
time to end its life.

One of the investigations by our team was to decide what the best robotic technology would be that we could readily fly in our mission and that would be most suitable for our application.  The two leading contenders were the Robonaut system from the NASA Johnson Space Center and the Special Purpose Dexterous Manipulator from MDRobotics in Canada.

Visit to the robonaut lab at the Johnson Space Center (click for more)

Due to our extremely short schedule, the SPDM robot from MDRobotics was chosen as the Dexterous Robot on our robotic servicing mission.
Having selected a robot arm, we performed feasibility test with mockups of Hubble hardware at MDRobotics in Canada working along with our team.  We hoped that these tests would show that the tasks formerly performed by humans could be done with manageable risk by a robotic system.

At MDR in Canada
Initial tests of installing the Diode Box and Advanced Computer connectors in Canada (click for more).

Showing success, the above tests were continued back home at the Goddard Space Flight Center.  At this phase, we handled mockups of science instruments and more tools.  Specifically, we removed and inserted a mockup of an instrument I have been working on called Wide Field Camera 3, and I became involved with the changeout of a large bulky instrument called the Cosmic Origins Spectrograph (COS).

COS mockup test
Tests with the COS mockup and the Capaciflector at GSFC (click for more)

These feasibility studies were so successful that it convinced the NASA Administrator and the rest of the agency that it would be possible to service the Hubble with a robotic system.  As mentioned above, we found out in October that we were approved to pursue the mission until our Critical Design Review in September 2005.  This put our project into high gear.  NASA considers this mission to be the first exploration mission in relation to the President's vision for the Exploration of the Universe.  This exploration will need large spacecrafts to be assembled in space using robotics controlled from the ground.  In addition to gaining useful experience with this activity, we have the chance to save the Hubble Space Telescope, the most useful space observatory ever.

One of our next activities was to meet with past Hubble Astronauts to benefit from their personal experience during their mission.

Crew TIM
Technical Interchange Meeting with past Hubble Astronauts at the Johnson Space Center (click for more)

Meanwhile, the number of newspaper articles were growing,  such as these:
Washington Post article (11/12/04)
 Chemical & Engineering News (11/29/04).

In December 2004, we proceeded with our System Requirements Reviews, and by
then the design matured to the concept shown above.  As you can see in the above figure,
the Solar Arrays have been added to the 'bottom' part to form a sun shield to prevent
the entry of solar energy into the exposed parts of the Telescope. 

The Hubble Robotic Vehicle consists of two parts.  One part, called the Ejection Module (EM), holds the robot arm, the new instruments that will be replaced, and the tools that are required.  Once servicing is complete, it will be ejected from the system, leaving only the Deorbit Module (DM) behind on the Hubble Space Telescope.  The DM is the white faceted object just below HST in the picture above.  By minimizing the amount of mass that is left behind in the DM, we can better allow HST to point and 'slew' across the sky.  The DM will contain a large rocket engine that will allow the HST to be 'deorbited' once its mission has been complete to ensure safe disposal of the Telescope in the Pacific Ocean.

Due to the cost of launching heavy items into space, robots used in a micro-G environment tend to be very flexible and are unable to lift themselves in the presence of gravity on the surface of the earth.  As a result, it is not feasible to train the robot operators with them, and we need to rely on computer simulations instead.  In order to learn about the current advances in simulation and training facilities, we visited the Canadian Space Agency, who has produced the robots used on the Shuttle and the Space Station.

csa visit
Visit to the Canadian Space Agency.  Click on image for more.

Back home at Goddard, we have set-up the Dexterous Robot Ground Trainer to
perform tasks and to test tools.

In March 2005, Miles O'Brien from CNN came to interview some of our team members.

In January 2005, we were dismayed to learn that about the
Decision by the White House to drop Hubble Servicing from 2006 budget (1/05)

Undaunted, we press on buoyed by:
Statement by Senator Barbara Mikulski supporting Hubble Servicing (3/05)
Statement by Congressman Steny Hoyer (3/30)

and by the National Academy of Science's
Final report for Assessment of Options to recommend the extension of  the life of Hubble (3/05)

Simulator Demo
On March 21 2005, we held our Preliminary Design Review (click for more).
The review was extremely successful, and we impressed the rest of NASA about
our ability to pull this mission off.  Nevertheless, current plans are to not perform a
robotic servicing mission.  Article on

A quote from Mark Borkowski (NASA's Program Executive for the HRSDM) :
"Last week’s preliminary design review for Hubble servicing was one of the better ones I’ve seen"
"a super job".
Article on

Fight continuous to save Hubble:
Article in (free registration needed).

On 4/29/2005, Dr. Mike Griffin, the NASA Administrator made the decision to stop development of the
Hubble Robotic Servicing and De-Orbit Mission, and resume work on a Shuttle Servicing Mission.
Transcript of press conference.

As covered in the press:
Baltimore Sun (subscription needed)
Bangor News
Houston Chronicle
Florida Today
Christian Science Monitor

This means stop work on the De-Orbit Module, and the Ejection Module. 
However, due to the progress we have made with the robotic system,
the Administrator has also decided to continue our group's work in the robotic area. 
The President's Vision for Space Exploration will need robotic technology,
and our group may have a role in that future.  As a result, I
am continuing my work with the Contact Dynamics Lab.

These days much of space related research and development is done in the private
sector by companies like SpaceX. Such private enterprises need to be knowledgeable
about IRC Section 41, which is the IRS code for calculating R&D tax credits.


On one of  our visits to Canada, we visited the CN Tower.

Other missions

Back Home

(c) Edward Cheung, 2005