At times, Frank Cepollina imagines what it would be like to retire from NASA, return to Northern California where he was raised, and relax. For now, however, Cepollina, one of the world’s leading experts on refueling and repairing satellites in orbit, has far too much work to do.

Cepollina, NASA deputy associate director in the Space Servicing Capabilities Office at NASA’s Goddard Space Flight Center in Greenbelt, Md., is preparing to send a mock spacecraft his team designed to the international space station in preparation for an experiment to evaluate the capability of robots to refuel and repair satellites in orbit. At the same time, Cepollina is planning a ground-based demonstration of the hardware and software needed to assist robotic spacecraft in approaching orbiting satellites and latching onto them.

Throughout his career, Cepollina, who joined NASA in 1963, helped pave the way for in-orbit satellite servicing. In the early 1970s, he designed and developed the Multimission Modular Spacecraft, which featured common components to provide power, attitude control and other capabilities. Cepollina later oversaw each of NASA’s campaigns to repair satellites in orbit, from the Solar Maximum Mission in 1984 to the most recent Hubble Space Telescope servicing mission in 2009.

Cepollina, who is 74, spoke recently with Space News correspondent Debra Werner.


Why should the United States develop the capability to repair and refuel satellites in orbit?

There are enough assets in space, especially in geosynchronous space, that it’s well worth the United States’ making an investment to preserve and extend the life of those assets.


What is the most challenging part of extending the life of those satellites?

All those satellites were designed without any plans for repair and maintenance in orbit. The other satellites we were able to repair and upgrade, Hubble Space Telescope, Solar Maximum Mission and Compton Gamma Ray Observatory, were designed to allow humans and robotic arms to grab them and service them. Now we are talking about spacecraft that were never designed to be robotically repaired and refueled. Is it possible to go up there with a robot, take off the safety wires and insulation, unscrew the safety caps, open the valves, and pump hydrazine and nitrogen tetroxide into the original tanks of those satellites? It’s never been done before in space.

We are talking about refueling, repairing, tugging satellites to the right location or tugging them outside the geosynchronous belt so they won’t be in the way.

Those are big challenges, but they are challenges we can meet if we take a serious, succinct, step-by-step approach. We are trying to do that.


What is the plan for the Robotic Refueling Mission?

We are launching it in June on Space Shuttle Atlantis’ STS-135 mission. The purpose is to demonstrate that we can do those tough refueling jobs. We will conduct a regular demonstration over 24 months. Robots controlled by personnel on the ground at [the Johnson Space Flight Center in] Houston will send commands to the robot on station equipped with special tools that we built. The robot will grab the right tools, do the job and flow fluid into a receiving tank.

That’s how we will learn all the engineering subtleties associated with actually doing the job so that by the time we refuel an actual spacecraft, we will have that engineering behind us. The demonstrations will occur, on average, about three to five days a month. There will be no interaction by astronauts on the station.

The second thing we will do is demonstrate replacement of several different kinds of boxes. Some satellites need repairs, reconnections, deployment of appendages and that sort of thing. We will conduct a series of experiments during that same 24-month period to demonstrate our ability to repair and replace boxes.


Is all the work focused on the satellite mock-up you developed at Goddard?

Yes. We put in all the smarts we could and, in some cases, satellite users said, “Hey if you work on my satellite, this is what you are going to encounter.” So we have different kinds of fuel valves, electrical connectors, RF connectors, cryogenic connectors, all different kinds of things geared toward our ability to demonstrate we can do this kind of work in space reliably.


What is the next step?

Most orbiting satellites don’t have any kind of grapple fixtures. Owners and builders said, “What makes you think you can find our satellites in geosynchronous orbit, grapple to them and control them?”

The second project is a set of ground-based experiments we call the Approach, Rendezvous and Capture Demonstration Test. We are doing that at Goddard, at the Naval Research Laboratory and at West Virginia University. We will test our software, test the robot’s ability to get close to a satellite, receive precision commands and orient itself so it’s perfectly aligned with the spacecraft. Then it will go in and make the capture. We are going to conduct those tests in the next 12 months.


Those two experiments will occur simultaneously?

Yes. And at the same time, we hope to be able to begin development of a refueling and repair spacecraft mission. It would be just like a tow truck on Route 66. We would carry fuel and repair capability to be able to deploy stuck appendages and that sort of thing.


That one you will send into space?

Yes. That’s about three-and-a-half, four years off, but we want to get started developing and designing that this calendar year. Within the next 12 to 18 months, we will have enough information from the demonstrations we are conducting to inform our designs.


Do you have a budget for that?

We have a budget, but I’d rather not mention any numbers.


Are other government agencies supporting this?

What we are really looking for besides government agencies is commercial partners. Commercialization is the end goal here. In the early 1960s, because we launched a couple of communications satellites into space, we became the catalyst for the commercial communications industry.


Can you identify any potential partners?

No. We are going to issue a request for information on that subject. We will ask companies with an interest to come and talk to us. And there are many, many different forms of commercial ventures. There are a lot of possibilities here.


Is the goal to create a commercial business?

Yes. Our first goal, first and foremost, is to be able to push robotic technology. We have a robot in the United States called da Vinci developed by Intuitive Surgical that can do very delicate surgical operations on humans. The surgeon operates the robot in a telerobotic manner, very similar to the way we would fix satellites. Now the degree of motion, the degree of dexterity of that robot is significantly different for medical operations than it would be for space operations. But what we want to do is push those kinds of technologies forward so you can do brain surgery in orbit. Not today or tomorrow but in 10 or 15 years. The potential for using that kind of technology for doing very precise repair activities becomes very valuable in terms of preserving and extending one’s use of national assets and private assets.

There are also strategic interests. The U.S. needs to be one of the leaders, if not the leader, for in-orbit robotic operations.



We have military satellites that never quite make it to orbit. Milstar-1 was a good example. That happens statistically in a regular mode. From a national, strategic point of view you want to make sure that you can preserve your capabilities for in-orbit communications and other operations. These are very expensive assets.


What do you think of plans by MacDonald Dettwiler and Associates (MDA) to refuel satellites for Intelsat and others?

My personal opinion is that it is wonderful more than one nation is interested in this endeavor. It shows that on an international level, the aerospace community is recognizing the economic benefits of being able to extend the lifetime of existing spacecrafts. As our allies advance in their capabilities, both national and international benefits will follow.

Relative to MDA, we’re helping each other through the Canadian Space Agency and international space station, but we’re not directly coupled. My goal is to develop the technology that will enable us to repair and refuel any spacecraft, at any time, in any orbit from low Earth orbit to Lagrange points 1 and 2. The first step is robotic servicing at geosynchronous orbit.


What do you want NASA, other government agencies and Congress to do to push in-orbit satellite servicing along?

Take interest and fund us. I have been working on satellite servicing since 1976. We have always gotten the support and backing that we needed. We did five Hubble repair missions ranging in complexity from mundane work to the last mission where we cut a hole in the middle of a scientific instrument, took circuit boards and plugged in new circuit boards. The instrument produced some amazing scientific breakthroughs after it was repaired.

Debra Werner is a correspondent for SpaceNews based in San Francisco. Debra earned a bachelor’s degree in communications from the University of California, Berkeley, and a master’s degree in Journalism from Northwestern University. She...