VIII
In-space infrastructure & servicing
Docking, refueling, debris capture and orbital transfer are pre-revenue but program-funded; the qualified bench is tiny.
In-space servicing is the chokepoint where the space economy's narrative is loudest and its revenue is quietest. The physical constraint is concrete: docking with an uncooperative client, transferring hydrazine or xenon, capturing tumbling debris, and relocating a satellite in geosynchronous orbit all require flight-qualified rendezvous-and-proximity-operations (RPO) hardware, robotic arms, and standardized fueling interfaces. Almost none of it is generating commercial revenue yet. The entire qualified bench fits on one page, and that scarcity — not a TAM slide — is the investable fact.
Who owns it
The shipped-hardware bench is tiny and largely government-funded. Northrop Grumman's SpaceLogistics has flown two Mission Extension Vehicles that docked with and are still flying Intelsat satellites, and its DARPA-backed Mission Robotic Vehicle (RSGS) — carrying robotic arms developed with the Naval Research Lab — is in test in Virginia for a summer-2026 Falcon 9 launch from Cape Canaveral. MDA Space holds the Canadarm heritage, a C$1B Canadarm3 contract inside a ~C$4.8B backlog, and a commercial Skymaker robotics line. Astroscale's US arm is building the APS-R refueler for the Space Force under a $61M OTA plus a $25.5M award, fueling itself from an Orbit Fab depot. Redwire (via Made In Space) built the Archinaut/OSAM-2 additive-manufacturing structures — a ground-concluded demo that did not reach orbit; the SPIDER robotic-assembly payload on OSAM-1 was Maxar's. Intuitive Machines is pushing a Nova-C-derived Orbital Transfer Vehicle through Critical Design Review. The refueling-interface standard layer is owned by Orbit Fab, whose RAFTI port is a Space Force-accepted interface.
What breaks it
The edge is that servicing requires three things SpaceX does not supply: RPO/docking flight heritage, dexterous robotics, and an agreed fueling standard. Heritage cannot be bought with capital — the MEVs are the first commercial GEO dockings ever flown, and the MRV uses NRL arms a decade in the making. The standard problem is the deeper choke: a refueler is worthless without ports on client satellites, which is why Orbit Fab spent years getting RAFTI accepted and why Astroscale's APS-R refuels via that depot architecture. Demand is real and program-funded (DARPA, Space Force, NASA, CSA), but it is government demand, not a commercial book.
What forces the reprice
This chokepoint reprices on flight events, not awards. The MRV's summer-2026 launch is the first US robotic GEO servicer to fly — though, on electric propulsion, it takes ~10 months to spiral up to GEO before it services anything. Astroscale's APS-R is also slated for summer 2026 to GEO, the first refueling of a US national-security asset. A successful docking or fuel transfer converts "funded program" into "demonstrated capability" and is what re-rates the bench from option value to backlog.
Be honest about size and earliness
This is pre-revenue dressed as a sector. Northrop's servicing line is rounding error inside a defense prime; you own NOC for the whole company, not the MRV. Redwire booked $97M in Q1 2026 revenue (+57.9%) but lost $76.5M, and a growing share of that revenue is now Edge Autonomy drones, not orbital servicing — a textbook space-washing trap if you tier it by the renders. Intuitive Machines' OTV is a $9.8M study contract; its $1.1B backlog is lunar landers, not servicing. Astroscale's flagship ELSA-M debris demo slipped to FY2028+ after its launcher failed, and its US shares trade thin on OTC. And SpaceX — which just debuted on Nasdaq June 12, 2026 at a >$2T market cap — IS the launch market, not a servicing play; it carries these servicers to orbit but does not compete in the choke. The denser day-to-day coverage (Payload, SpaceNews) maps the news; this thesis maps the supply chain: who has docked, who owns the fueling standard, and which flight in the next two quarters turns a render into a reference mission. This is structure and catalysts, not advice.
Who owns the choke
Astroscale Holdings Inc.
Astroscale's US arm is building the APS-R refueler to transfer hydrazine to two US Space Force GEO spacecraft, funded by a $61M OTA agreement plus a $25.5M award and slated to launch summer 2026 using an Orbit Fab propellant depot; its commercial ELSA-M debris-removal demo slipped to fiscal 2028 or later after its Isar Aerospace launcher failed. Trades OTC in the US as ASRHF; primary listing is Tokyo Stock Exchange Growth (186A).
- Bull
- If on-orbit refueling becomes real, Astroscale's US arm has a funded $61M-plus Space Force APS-R mission slated to transfer hydrazine to two GEO craft in summer 2026.
- Bear
- Its commercial ELSA-M debris demo already slipped to fiscal 2028+ after a launcher failure, and the US line trades thin OTC with the primary listing in Tokyo.
Intuitive Machines, Inc.
Intuitive Machines is advancing a Nova-C-derived Orbital Transfer Vehicle for high-energy-orbit and cislunar payload delivery through Critical Design Review under a $9.8M Phase Two contract, targeting manufacturing/flight integration as early as 2026. Reported a record $1.1B backlog in Q1 2026, driven mainly by lunar-lander (CLPS) and the Lanteris acquisition.
- Bull
- If in-space transfer becomes real, Intuitive Machines' Nova-C-derived OTV is advancing through Critical Design Review toward flight readiness as early as 2026.
- Bear
- The OTV is only a $9.8M study contract; the record $1.1B backlog is lunar landers and the Lanteris acquisition, not servicing revenue.
MDA Space Ltd.
MDA holds Canadarm heritage and a C$1B Canadarm3 flight-system contract (Phase C/D) for NASA's Lunar Gateway inside a ~C$4.8B backlog; its Skymaker robotics line supplies arms for commercial stations including Starlab. MDA listed on the NYSE under symbol MDA via a ~US$300M US IPO that closed March 16, 2026 at US$30.50/share.
- Bull
- If servicing reprices on robotics heritage, MDA holds Canadarm lineage, a C$1B Canadarm3 contract inside ~C$4.8B backlog, and the commercial Skymaker line.
- Bear
- As a newly NYSE-listed name (March 2026 IPO at US$30.50), MDA carries execution risk on long-dated government programs and limited US trading history.
Northrop Grumman Corporation
Invested $50M in Firefly in May 2025 to co-develop the Eclipse medium launch vehicle (16.3t to LEO, Firefly Miranda engines and composite structures), an outgrowth of the Antares 330 collaboration (Firefly's >$2B Series D closed Nov 2024). Northrop also supplies GEM solid rocket boosters used on ULA's Atlas/Vulcan.
- Bull
- If servicing and reusable launch advance, Northrop's two flown MEVs and summer-2026 DARPA MRV give servicing heritage, plus a $50M Firefly Eclipse stake.
- Bear
- The servicing line is rounding error inside a defense prime; you own Northrop for the whole company, not the pre-revenue MRV exposure.
Orbit Fab, Inc.
Orbit Fab's RAFTI (Rapidly Attachable Fluid Transfer Interface) refueling port is an accepted refueling interface for US military satellites and underpins the depot Astroscale's APS-R will use; the company has raised over $30M including a $28.5M Series A. Private; US-listed exposure is only indirect via strategic investor Lockheed Martin's venture arm.
- Bull
- If refueling needs a standard, Orbit Fab's RAFTI port is a Space Force-accepted interface underpinning Astroscale's APS-R depot, having raised over $30M.
- Bear
- Orbit Fab is private; US-listed exposure is only indirect via strategic investors Lockheed Martin and Northrop, so there is no direct tradable position.
Redwire Corporation
Redwire is co-developing the Valkyrie Hall-effect thruster (based on NASA's H71M design) in a strategic JV with private propulsion firm Phase Four, targeting high-volume DoD smallsat supply; the company has cited up to $150M annual revenue potential, which is a market-opportunity figure rather than booked backlog. Redwire reoriented in 2025 toward defense tech via the Edge Autonomy acquisition.
- Bull
- If smallsat propulsion scales, Redwire's Valkyrie Hall-effect JV with Phase Four targets high-volume DoD supply, with cited up to $150M annual revenue potential.
- Bear
- The $150M is a market-opportunity figure, not backlog; Redwire's revenue increasingly reflects Edge Autonomy drones, a textbook space-washing dilution of the thesis.
Catalyst calendar
- 2026-08-12Redwire Q2 2026 earningsmediumTests whether orbital servicing/assembly (OSAM, SPIDER) revenue is growing or whether the story is increasingly Edge Autonomy drones — the key space-washing read on this name.
- 2026-08-15Northrop Grumman / DARPA Mission Robotic Vehicle (RSGS) launch on Falcon 9mediumFirst US robotic GEO servicer to fly; a successful launch and deployment converts the in-space servicing bench from funded program to demonstrated capability.
- 2026-08-18Intuitive Machines Q2 2026 earningsmediumWatch for Orbital Transfer Vehicle progress past CDR toward a funded flight order, which would move OTV from optionality to backlog in the orbital-transfer leg of the chokepoint.
- 2026-09-15Astroscale APS-R Space Force refueler launch to GEOmediumFirst refueling of a US national-security asset and the first flight test of the RAFTI/Orbit Fab depot fueling-standard architecture that the whole refueling choke depends on.