With the successful docking of Mission Extension Vehicle 2, or MEV-2, to the Intelsat 10-02 satellite last month, Northrop Grumman not only repeated the task of successfully attaching one of their MEV spacecraft to a functioning satellite but also successfully proved the ability to grab a still-transmitting telecommunications satellite without disrupting service.
The success of both MEV-1 and -2 has led to an increasing interest in the use of those crafts after their current five-year missions with their present satellites are complete. Meanwhile, Northrop Grumman has already begun work on the next generations of remote, on-orbit servicing and debris clean-up vehicles.
MEV-2 builds on MEV-1’s success
Launched in October 2019, MEV-1 rendezvoused with its target satellite, Intelsat 901, on 25 February 2020, successfully performing an automated rendezvous and docking in an area of Earth orbit known as the GEO graveyard.
The GEO graveyard is located approximately 300 kilometers above Geostationary orbit, which itself resides at 35,786 km above Earth sea level.
The first-ever docking in this type of Earth orbit, MEV-1 successfully demonstrated the ability to grab a still functioning but not transmitting or operational-in-that-regard satellite and provide mission extension propulsion and attitude control services.
MEV-1 successfully maneuvered Intelsat 901 back down into the operational GEO belt, allowing it to continue to use its still operational telecommunications services even though its onboard propulsion system was running out of fuel to keep the satellite stable in orbit.
Today, we successfully completed the docking of @NorthropGrumman’s Mission Extension Vehicle-2 (#MEV2) to Intelsat 10-02 in order to extend its life, proving once again that in-orbit servicing can be done. Learn more: https://t.co/ESmJTa6Fx2 pic.twitter.com/NuuvVQWiZa
— Intelsat (@INTELSAT) April 12, 2021
Building on the success of MEV-1, MEV-2 successfully launched in August 2020 on an Ariane 5 ride-share mission into Geostationary transfer orbit. It then spent the months after launch slowly raising its orbit up to GEO altitude inside GEO’s operational area assigned to its target satellite – Intelsat 10-02.
Therein is the first major difference between the two missions. MEV-2 was not grabbing a non-operational but still functioning satellite; it was instead given the obligation of docking to a still-transmitting telecommunications satellite in Geostationary orbit.
In this case, going directly to the target satellite while it was still operational in some ways simplified the operations of getting MEV-2 to the correct point in space where it was ready to dock to Intelsat 10-02.
According to Joe Anderson, Director, Mission Extension Vehicle Services, Northrop Grumman, in an interview with NASASpaceflight, “Docking on MEV-1 in the graveyard orbit, we had to use a lot of special operations to avoid [Radio Frequency] interference with other operating satellites in GEO as we were drifting past them.”

Intelsat 10-02 seen from MEV-2 during the latter’s hold during approach at the 15-meter Waypoint ahead of docking on 12 April 2021. (Credit: Northrop Grumman)
“MEV-2 was a little bit simpler for us because we didn’t have that; we weren’t drifting past other satellites.”
Something from MEV-1 that was not originally planned for inclusion on MEV-2’s mission but proved so useful with MEV-1 that Northrop Grumman decided to make it a normal procedure was a calibration — or practice — approach prior to the actual docking.
“On MEV-1, we had incorporated something we called a calibration approach. Because it was the first time, we wanted to do a practice approach to the client and make sure all our sensors were tuned up properly and that all the systems on both the client’s satellite and our satellite behaved properly as we got close,” said Anderson.
“We found, actually, that that was a really good idea. Originally, we didn’t intend to continue that on our subsequent dockings. But based on what we learned there, we decided that that’s something we definitely wanted to incorporate into our future missions as well.”
Another key change with MEV-2, and a lesson learned from MEV-1, was the addition of a Waypoint, or location along the approach vector where the MEV stops to ensure it is properly aligned with its docking target on the client satellite.
For MEV-1, three Waypoints were used, one at 80 meters distance, one at 15 meters, and the final at 1 meter, at which point the docking sequence was carried out.
“What we found from that,” explained Anderson, “is that it would improve our performance and our confidence in our alignment for the docking if we were to add another waypoint about 3 meters behind the client.”
The new Waypoint was employed on MEV-2’s approach to Intelsat 10-02 and allowed for better control of the actual docking timing given the satellite would still be transmitting to customers on the ground. The new Waypoint also allowed better confirmation of alignment with the liquid apogee engine on the back of Intelsat 10-02, which was MEV-2’s docking target.
“Intelsat wanted to establish a service window for their customers. Their customers knew when they might expect a disturbance in their traffic,” noted Anderson.
However, that never happened.
“[Adding that Waypoint], that was a good decision. It really paid off for us on MEV-2, as when we did dock, we had zero transients. We had no customer outages. None of Intelsat’s customers experienced an outage when we docked.”
Docking was conducted in the same manner used for MEV-1, with a docking probe on MEV-2 extended into the liquid apogee engine on Intelsat 10-02. Once the docking probe passed the smallest part of the nozzle opening, known as the throat, the probe expanded and, like a wall anchor, provided a secure way to slowly pull Intelsat 10-02 down onto the docking clamps of MEV-2, which themselves attached to Intelsat 10-02’s launch adapter ring.
The method for docking an MEV with a satellite that was never designed to be docked to or serviced in space is a careful part of the overall Mission Extension Vehicle design.
“The key there is really finding those features that are present on a large number of GEO satellites that we could attach to because we’re docking to satellites that were not designed to be docked with or serviced,” noted Anderson. “There are two key factors that are present at about 80% of all of the satellites in GEO. That is a liquid apogee engine and a launch adapter ring.”
The launch adapter ring is no longer needed once the satellite separates from the rocket’s upper stage that launched it. The liquid apogee engine is only used for the initial orbit-raising maneuvers to begin the process of getting the satellite into a proper geostationary orbit after launch.

Intelsat 10-02, seen from MEV-2 while the latter was approximately 80 meters behind the satellite on 12 April 2021. (Credit: Northrop Grumman)
Additionally, the MEVs have to be able to dock to satellites using different buses. These different buses have different properties that affect automated rendezvous and docking operations, such as reflectivity, orientation of solar panels, and placement of attitude control thrusters.
In fact, even though MEV-1 and MEV-2 both docked with Intelsat satellites, Intelsat 901 and 10-02 use completely different buses, which had to be accounted for when MEV-2 approached its target.
As Anderson related, “The client satellites for MEV-1 and MEV-2 are two different satellite buses. One was made by Space Systems/Loral at the time, Maxar now, and the other by Airbus. Those satellites each have their own particular features. They look different, they have different reflective properties, they have different ways that they do their attitude control, and so you have to be very careful about accounting for all of those as you do your rendezvous approach and docking.”
Success and future
The success of the MEV program so far has certainly been seen throughout industry, with interest growing from potential clients.
“After MEV-1, we received a lot of calls. ‘Can I get that MEV next?’ ‘Can I get it now?’ ‘If we have a problem, is there any way I could use it?’ ‘MEV-2 is coming, can I get MEV-2?’ We got a lot of interest like that.”
“I’ve been saying for quite some time that this market is a ‘build it and they will come’ type of market. We’ve seen good evidence of that since I started working on this in 2012 and visiting customers.”
In particular, Anderson noted interest within the community as far back as 2012; however, a major hesitation from customers was due to their need for such services immediately while not having a way to adequately predict what their needs would be three, four, or five years later.
Anderson found that as the years passed, potential customers would continue to say they required the service right then… but those specific needs changed from year to year.
“That was the first evidence of: if we build it, if we are there in orbit, those customers will be there,” said Anderson. “There is just this latent demand for this type of service.”
But in all of those yearly and regular conversations where Anderson sussed out what the changing needs of customers were, a pattern clearly emerged. There was a large need for different types of robotic, automated servicing missions for perfectly fine and still operational satellites that were simply running out of fuel to continue to be able to point in the correct direction for service as well as to maintain the orbits needed for those operations.
In part, this has led to the development of not just the next generation beyond the MEVs but the next generation beyond the next generation, so to speak, of automated, geostationary orbiting servicing fleets.
“First, we have our next generation system that we’re already constructing. It’s called our Mission Robotic Vehicle and that’s done in a partnership with DARPA, where DARPA is providing the robotics system.”
Basically a mini-MEV, these Mission Robotic Vehicles will be able to move from satellite to satellite in Geostationary orbit installing propulsion augmentation systems called mission pods, to satellites like Intelsat 901 and 10-02 that are still functioning but simply running out of propellant for attitude and/or orbital control.
The mission pods would provide six years of mission extension service in the form of attitude control.

Artist’s depiction of a Mission Robotic Vehicle holding a mission pod. (Credit: Northrop Grumman)
After attaching the mission pods, the Mission Robotic Vehicle (MRV) would undock and move off on another mission. In addition to attaching mission extension pods, the MVRs would be able to grab satellites and move them into different orbits as well as assist with debris clean-up activities in GEO.
“We are doing studies into the feasibility of using that robotic vehicle to grapple debris in the GEO orbits,” noted Anderson. “There is some debris there. It’s not a huge problem in GEO, but there are some cases where customers would be very interested in having a piece of debris removed. We are looking at and evaluating the feasibility of doing those types of missions out in the GEO belt.”
This opens the possibility that the technology employed on the MRVs could be used for other debris cleanup operations, specifically the more cluttered low Earth orbit environment.
“All of this technology could be applied to those types of debris removal problems,” said Anderson. “Now the issue that we see with it right now is there is no customer base. There is no one right now that is incentivized to pay for those types of services.”

A mission pod attached to a client satellite. (Credit: Northrop Grumman)
But even beyond that, the third generation of robotic servicing vehicles are already in the planning stages, as well as how they will integrate with future satellites launched towards geostationary orbit.
“We’re already starting our generation three, a third generation of GEO servicing for refueling of prepared satellites,” related Anderson.
“Our approach is to start doing refueling with satellites that are prepared for refueling. We’re developing refueling interfaces that we would like to make an open industry standard. Then our vision here is that by 2025, every new satellite that is launched is prepared for servicing in some way.”
This third generation of vehicle would not just be able to perform refueling operations but also robotic servicing as well using robotic arm technology to repair elements on the exterior, or even interior, of satellites — including an ability to remove and replace solar arrays.
“Designing solar arrays so they can be taken off or put back on or add additional solar arrays to it… absolutely, that’s on the roadmap,” enthused Anderson. “That really gets to the next step of our roadmap, actually. Beyond satellites prepared for servicing is in-space manufacturing, in-space assembly of spacecraft.”
“That’s something we see coming. There’ll be a lot of development and incremental capabilities of that over this decade, but we think it really starts to become a capability that we can utilize in the 2030s and beyond.”
(Lead image: Artist’s impression of an MEV docked to a client vehicle in GEO. Credit: Northrop Grumman)