Debris Removal is the Cover Charge: How the USG is Bankrolling a Domestic RPOD Capability
Oct. 8, 2026
By Kimberly Siversen Burke
Katalyst's first servicing spacecraft reentered the atmosphere on Sept. 25 without boosting NASA's Swift Observatory to a higher altitude as planned. An attitude-control failure during commissioning forced NASA and the Colorado-based defense tech startup to abandon the capture, ending the LINK spacecraft's mission after 85 days in orbit.
Six days later, Starfish Space launched its first full-size Otter servicer, Otter 24C, aboard SpaceX's Transporter-18. The 335 kg spacecraft is built to dock with satellites that were never designed to be serviced, whether for life extension or disposal. Under NASA's Small Spacecraft Propulsion and Inspection Capability (SSPICY) mission, it will inspect up to four dead U.S. satellites and rocket bodies starting in 2027.
Otter 24C was preceded by a smaller but arguably more famous Otter Pup. In 2023, Starfish had the space internet refreshing its feeds while it nursed a tumbling Otter Pup back from “the brink of death.” And just this week, Starfish’s Otter Pup 2 is writing the sequel. As of Oct. 6, Pup 2 was holding about 1 km from Gilmour Space's ElaraSat, steering on its own autonomous navigation ahead of a docking attempt. Next up is the Otter that the Space Development Agency (SDA) has under contract that will fly Starfish's first disposal mission in 2027.
Both SDA and the Defense Innovation Unit (DIU) have become the Pentagon's frontrunners in awarding deorbit contracts in an unsung effort to cultivate a U.S. rendezvous, proximity operations, and docking (RPOD) capability ahead of any meaningful commercial market traction.
Deorbiting is the shallow end of orbital mobility because the target is already dead, so a botched docking attempt doesn't threaten a working satellite. SDA can justify the mission as disposal for now, while vendors use these small(ish) awards to develop the capability the Pentagon is ultimately after – a spacecraft that can refuel, repair, and reposition operational satellites on orbit.
When Satellites Play Hard to Get
SDA is leaning on DIU – the Pentagon’s Silicon Valley-born shortcut to commercial technology – to build a roster of contractors that can capture and deorbit dying satellites on demand. DIU runs the Deorbit-as-a-Service program for SDA under a managed-service model, with the contractor owning and operating the servicer. But last month's reentry exposed the gap between what those companies have demonstrated and what the government expects them to deliver.
NASA hired Katalyst in September 2025 for a $30M mission to raise Swift's orbit. Swift is what DIU's solicitation calls an “unprepared” satellite, meaning it has no standardized docking interface for a third party to use. Despite being a third of Swift's size, LINK had to rely exclusively on onboard sensors to track the massive observatory, calculate its exact attitude, match its orbital velocity, and execute a soft capture without damaging either vehicle.
But LINK never got the chance.
An electrical fault knocked out two of its three reaction wheels – the spinning flywheels that point the spacecraft. Katalyst switched to firing thrusters to hold LINK's orientation, and that burned through the propellant reserved for the capture, forcing NASA and Katalyst to call off the attempt on Aug. 19. LINK got as close as 12–15 km from Swift, but it didn't have enough fuel left to finish the job.
The Swift mission (an orbit raise, not a disposal) is a reminder that deorbiting alone probably won't be enough to underwrite these companies' business models. At least for now, each vehicle must earn its keep across a range of missions, from disposal to inspection, repositioning, and life extension. So the spacecraft’s commercial viability depends on it being more of a Swiss Army knife than a garbage truck.
Katalyst is already hedging that bet. The space robotics company is carrying LINK's flight data into NEXUS, the multi-mission spacecraft it's planning to use for its Deorbit-as-a-Service contract with DIU for SDA. And NEXUS already has another job with the Pentagon lined up. The Operational Energy Capability Improvement Fund (OECIF) inked a deal with Katalyst on Sept. 21 to fly NEXUS on the Advanced Scalable Growth Architecture for Robotic Deployment (ASGARD) program. The demo will robotically assemble Naval Research Laboratory power-beaming tiles in LEO and beam solar power to the ground as microwaves for potential use at remote military sites.
But Katalyst isn't the only company treating disposal as one pitstop on a much longer road.
Nasdaq-listed Firefly Aerospace is building Elytra on hardware it has already flown. The spacecraft carries avionics, composite structures, and propulsion from Firefly's rockets and from its Blue Ghost lunar lander, which touched down on the Moon in 2025. Under its Deorbit-as-a-Service contract with DIU and SDA, Elytra would rendezvous with a dead SDA satellite, survey it, and deorbit it using a system that Firefly designed to handle satellites of very different sizes in LEO, MEO, or GEO.
D-Orbit Space, the Boulder, Colorado-based U.S. arm of Italian in-space logistics company D-Orbit, designed its space tug to hang out in orbit and service or deorbit multiple targets. The company is pitching its demo as a path to refueling, repair, inspection, and repositioning work for the government.
KMI Space, a Michigan startup that won a Space Force Strategic Funding Increase award in August, is taking a similar approach with Laelaps, a reusable spacecraft designed to rendezvous with, capture, and relocate multiple satellites. The name comes from the Greek mythological hound that never missed its quarry – an adorably ambitious choice for a 180 kg spacecraft.
Why Commercial Operators Are Sitting This One Out
Mission diversity is a key survival strategy for these companies right now, as the commercial RPOD market is primarily focused on extracting value through GEO life extension. Removing a dead satellite from LEO mostly buys reduced liability and compliance with disposal rules, and compliance just became easier to certify.
Under licensing rules the FCC adopted on July 22, operators now only certify that satellites are “designed and operated” to deorbit within five years of mission end, rather than guaranteeing it.
So the economics are brutal.
Tokyo-listed Astroscale, which holds JAXA's removal contract, reported a ¥36.4B backlog as of July, but its commercial customers accounted for just ¥391M — about 1%. Even its commercial demo (ELSA-M, targeting a Eutelsat OneWeb satellite) is drawing €13.95M from the UK Space Agency and ESA.
Disposal is an expense satellite operators try to avoid through spacecraft design or lower orbits where physics does its thing.
Look at SpaceX, whose satellites accounted for 75% of all active maneuverable satellites on orbit as of March 2026. Just this year, the company proactively began lowering about 4,400 Starlinks from 550 km to 480 km to slash a dead satellite's reentry time from more than four years to a few months during solar minimum.
But lower orbits and relying on gravity and drag for cleanup will not help SDA. Its satellites fly at about 1,000 km, where a dead spacecraft can keep circling Earth for a thousand years or more.
The Propulsion Crisis and SDA's $63M Solution
SDA's interest in RPOD started as a cost-savings measure, with then-Director Derek Tournear arguing in 2023 that he could just hire a commercial “trashman” to deorbit dead satellites instead of buying redundant propulsion.
The comment aged well as news of SDA’s propulsion supplier woes became more of a disAstra for the primes. In Q1 2024, Terran Orbital took a $13.1M charge after dropping its thruster supplier on the 42 buses it was building for Lockheed Martin's Tranche 1 Transport Layer.
So, in September 2024, SDA paid $1.9M for six disposal studies. Fast forward to 2026, and the agency awarded Starfish $52.5M for one deorbit, with options for more. This past August, DIU and SDA together added a reported $8.4M for design work from D-Orbit, Firefly Aerospace, and Katalyst. SDA has now cobbled together about $63M to fund a capability its vendors have yet to prove on orbit.
The point SDA is trying to make isn’t so much that disposal is valuable as much as self-disposal is expensive. Tournear's original pitch made sense: take the redundant propulsion off each satellite, accept that some small percentage will fail, and call a commercial servicer when they do.
U.S. Hires the Tow Truck; Europe Installs the Hitch
While the U.S. is trying to build servicers that can handle even the most uncooperative targets (very on brand for the U.S.), Europe is taking the opposite tack and making future satellites easier to approach and dock with before they even launch.
Since 2023, ESA has required its own new satellites in crowded orbits to carry fittings for servicers. Its June specification adds a standardized mechanical docking fixture and navigation markers to help a servicer locate and approach the spacecraft. The proposed EU Space Act would extend that idea, requiring EU-owned satellites to carry Spacecraft Service Interfaces starting in 2034.
Most of the world is still funding one-off removal missions:
Japan: JAXA is paying Astroscale ¥12.0B to deorbit a 3-ton H-IIA upper stage (targeting 2027-28).
UK: The UK Space Agency tendered a £75.6M mission to remove two defunct satellites by 2028.
ESA: The OHB-led ClearSpace-1 is targeting ESA's PROBA-1 satellite in 2029.
France: CNES hired Exotrail (with Astroscale France) to deorbit a OneWeb satellite in 2029–2030 (Astroscale's share: €13.2M).
Proven in GEO, Untested in LEO
The closest precedents for SDA's vision are still in GEO.
Northrop Grumman's MEV-1 docked with Intelsat 901 in February 2020, taking over its propulsion and returning the aging satellite to service.
China put the same capability to a very different use in 2022, when Shijian-21 docked with the dead BeiDou-2 G2 and hauled it about 3,000 km above the GEO belt before releasing it into a graveyard orbit.
The two missions bracket the range of what an on-orbit servicer can do. One kept a valuable satellite working while the other pushed a defunct one out of the way. SDA is trying to develop that same underlying capability for LEO, with deorbiting as the initial use case.
In LEO, NYSE-listed MDA Space already has more capture experience than any of SDA's vendors. Its flight controllers have conducted more than 100 free-flyer captures, most visibly the grapple-equipped cargo vehicles that Canadarm2, which MDA built, berths at the ISS. Now the space robotics and satellite manufacturing veteran is packaging that heritage into a spacecraft built for broader on-orbit missions. In April, MDA unveiled MIDNIGHT, a space-control platform pairing its SKYMAKER robotic arm with the AURORA bus. MIDNIGHT's first mission is to rendezvous with partner satellites in LEO, capture and release them, and deorbit a customer's non-operational spacecraft. MDA is courting military partners for that mission, while separately pushing AURORA into the defense sector with the AURORA BLACK spacecraft variant introduced last month.
The Space Force is Picking its Team
After preliminary design reviews, DIU and SDA will advance one Deorbit-as-a-Service design to an on-orbit demo.
The Space Force's Objective Force 2040 calls for operational “augmented maneuver” space tugs by 2030, and its Future Operating Environment 2040 treats servicing as a way to refuel and repair satellites. But neither document mentions standing debris removal.
Starfish already holds the closest thing to a government franchise with SDA's only disposal service contract, a $54.5M Space Force contract for a dedicated Otter, and NASA's inspection mission.
But Firefly is the only Deorbit-as-a-Service bidder that builds its own rockets, launched a satellite with ~24 hours' notice, and landed on the Moon. DIU's solicitation scores bidders on cost per deorbit, launch included, and on how quickly a servicer can reach orbit and start working. So, if the Pentagon wants one vendor accountable from the launch pad to the deorbit burn, that could be Firefly's game to win.
In launch, SpaceX went from plaintiff to pillar of U.S. national security in about a decade. Minus the bid protests and online rants, there is a useful lesson in that evolution.
In an emerging sector like RPOD, it’s a short runway for U.S. players not on the Pentagon’s roster. SDA’s commercial viability requirement is a sensible guardrail, but it shouldn’t be built before the road is paved.
###
