How Much Does It Cost to Rescue a Satellite in Space?

Published on July 25, 2026 | Written by Alec Pow
This article was researched using 12 sources. See our methodology and corrections policy.

Rescuing a satellite in space can cost about $30 million for a focused robotic orbit boost, but a custom refueling, repair, or crewed servicing program can reach hundreds of millions or several billion dollars. The mission price changes with orbit, spacecraft condition, docking access, launch requirements, development time, and the amount of new technology that must be built for one target.

The bill may combine a servicing spacecraft, capture hardware, propulsion, navigation software, launch services, environmental testing, mission control, tracking, licensing, insurance, and contingency funds. Commercial providers rarely publish full contract values, particularly for revenue-producing communications satellites, so federal awards, audits, and mission records provide the strongest public cost markers.

NASA’s Swift contract shows the price of a rapid orbit-raising attempt. SpaceLogistics and Intelsat show how commercial Mission Extension Vehicles can attach to fuel-depleted geostationary satellites and preserve operating life. OSAM-1 shows the budget risk attached to robotic refueling, autonomous rendezvous, custom tools, and spacecraft assembly in one program.

Hubble supplies historical crewed-servicing comparisons, where astronaut training and shuttle operations added costs that do not appear in robotic missions. GAO and NASA’s Office of Inspector General document schedule delays, contractor performance, technical uncertainty, and program reserves. Together, these entities connect the headline mission price to the work being purchased, the spacecraft being protected, and the financial value of keeping that asset in service.

How Much Does It Cost to Rescue a Satellite in Space?

Jump to sections

• Entry $30 million (at $30 per hour, earning that amount would take about 481 years of full-time work, before taxes) for NASA’s Swift orbit-boost award in September 2025.
• Historical servicing $136 million for Hubble Servicing Mission 3A in late-1990s dollars.
• Large crewed estimate $1.7 billion to $2.4 billion for a proposed Hubble shuttle servicing mission reviewed in 2004.
• Custom robotic program $2.1 billion to $2.17 billion estimated at completion for OSAM-1 in July 2023 before cancellation.

A satellite rescue may cost tens of millions of dollars for a limited orbit boost and hundreds of millions or billions for a custom repair, refueling, or crewed servicing program.

Satellite rescue is purchased per mission rather than per hour. The cost unit may be one target, one launch, one servicing spacecraft, or several years of attached propulsion support, with orbit class, target stability, capture hardware, and flight retries changing the final amount.

The Swift rescue award

NASA awarded Katalyst Space Technologies $30 million (about 481 years of full-time work at $30 per hour) in September 2025 to build, test, launch, and operate a robotic spacecraft intended to raise the orbit of the Swift Observatory. The published award does not divide that sum into vendor line items, so a buyer cannot treat the following planning split as Katalyst’s actual quote. The mission moved from contract award to launch in less than a year, making schedule pressure part of the purchased work rather than a separate administrative concern.

A mission planner could allocate 30% to spacecraft and capture hardware, 20% to engineering and software, 25% to launch and integration, 15% to flight operations, and 10% to reserve. Applied to $30 million, that produces $9 million + $6 million + $7.5 million + $4.5 million + $3 million = $30 million. This calculation is a planning model derived from the cited contract total, not a disclosed vendor invoice. It shows how launch and one-off hardware can consume $16.5 million, or 55% of the modeled budget, before engineering, operations, and reserve are funded.

What this is in plain terms

A satellite rescue is a mission that changes the condition, orbit, or remaining service life of a spacecraft already beyond Earth. The servicing vehicle may inspect the target, attach to it, take over attitude control, add propulsion, move it to a safer altitude, replace a component, or guide it toward disposal. Operators pursue this work when the existing payload can still deliver communications, weather data, navigation support, defense coverage, or scientific observations. The target remains the asset being protected, and the rescue craft acts as a temporary repair vehicle, propulsion source, tow vehicle, or permanent attached module.

The work differs from launching a replacement because the original spacecraft remains part of the operating system. It also differs from debris removal, whose purpose is disposal rather than restored use. Northrop Grumman’s Mission Extension Vehicle docks with a geostationary satellite and supplies propulsion after the client has depleted much of its own fuel. A robotic repair assignment may need cameras, autonomous navigation, manipulators, and custom tools. A crewed servicing flight adds life-support systems, astronaut training, shuttle or capsule operations, and human-safety controls. Those are separate service classes, even when each is described as a rescue.

Expenses across the mission

The first expense is mission engineering. Teams model the target’s orbit, rotation, communication status, docking geometry, structural limits, and remaining power. They then design approach corridors, retreat maneuvers, capture sequences, fault responses, and the final orbit change. A satellite built with a servicing fixture can accept a planned attachment method. An older spacecraft may require a mechanism that grips an engine nozzle, launch adapter, or another structural feature without damaging nearby equipment. Engineers must also account for solar arrays, antennas, thermal blankets, thruster plumes, and components that were never designed to support an external vehicle.

The next group covers hardware and delivery. It can include the servicing bus, propulsion system, robotic arms, cameras, lidar, computers, radios, capture tools, launch adapter, environmental tests, and rocket space. Operations begin before launch and continue through orbit checkout, rendezvous, inspection, capture, repositioning, and disposal. NASA’s 2023 audit found that OSAM-1 required added agency labor and technical support after contractor delays, helping push its estimated completion cost to $2.1 billion to $2.17 billion (about 34 to 35 thousand years of full-time work at $30 per hour), compared with a congressional commitment of $2.05 billion. That record shows why labor cannot be separated cleanly from hardware when late components force government teams to perform added integration and troubleshooting.

Mission element	What the buyer receives	Main price pressure
Engineering	Orbit analysis, capture design, software, testing	Unknown target condition
Servicing vehicle	Propulsion, sensors, robotics, communications	Custom hardware
Launch	Rocket capacity, integration, range support	Dedicated versus shared flight
Operations	Tracking, command, approach, docking, orbit change	Mission duration and retries
Risk funding	Schedule reserve, anomaly work, extra testing	Uncooperative target

Spacecraft, launch, and flight phase

Providers rarely sell satellite rescue through a public hourly rate card. The useful units are per servicing spacecraft, per launch, per target, and per period of attached operation. A mission-extension vehicle can serve one client for years, then detach and move to another. Northrop Grumman reports that MEV-1 completed a five-year client assignment with Intelsat 901 in April 2025 and docked with its next client during May 2025. One spacecraft can produce value across separate contracts, but each relocation still consumes propellant, navigation work, tracking capacity, and operating time.

Launch is another unit. A small rescue craft may fly as a secondary payload, occupy a dedicated small rocket, or share a larger mission. Each option changes schedule control, allowable mass, destination, and integration work. Flight operations may be budgeted by campaign length, but approach attempts cannot be priced like ordinary staff shifts. Each attempt consumes propellant, tracking time, navigation analysis, collision-risk review, and part of the vehicle’s remaining service life. Buyers should request separate figures for nonrecurring engineering, flight hardware, launch, mission control, attached service duration, undocking, relocation, and disposal. That format exposes whether a low headline figure leaves the rocket, insurance, licensing, or extended operations outside the contract.

Three rescue cases

Focused orbit boost. NASA’s Swift project represents the lower public end of the documented market. The agency committed $30 million under an accelerated schedule for robotic capture and an altitude increase. Its main driver is speed. Swift was not designed for servicing, and orbital decay created a deadline, yet the assignment stops short of opening the telescope or replacing instruments. Readers comparing the rescue with a new spacecraft can also review the separate economics of building and launching a satellite. The rescue case protects an operating observatory without funding an entirely new science platform, payload, launch campaign, and commissioning period.

Commercial life extension. MEV missions attach propulsion to fuel-depleted geostationary communications satellites. Contract prices are private, so a defensible public total is unavailable. The commercial value comes from keeping transponders active without waiting for a replacement. Intelsat extended its mission-support relationship for at least four more years in 2024. The primary cost driver is service duration rather than emergency speed. A provider must price the years of propulsion and attitude control supplied to the client, the fuel retained for later undocking, and the ability to move the vehicle to another satellite.

Custom robotic refueling. OSAM-1 sought to refuel Landsat 7, which lacked a purpose-built servicing interface, and demonstrate robotic assembly. An independent project review cited a $2.25 billion estimate in early 2024 before NASA ended the project. The gap between Swift’s $30 million award and OSAM-1’s $2.25 billion estimate is $2.22 billion. That difference reflects scope, development history, hardware complexity, program duration, and the need to demonstrate several capabilities rather than a standard market markup.

Orbit, access, and spacecraft condition

Low Earth orbit may sound easier because it is closer, but atmospheric drag and short visibility periods can compress the schedule. Geostationary orbit requires far greater travel energy, yet communications satellites there hold predictable operating positions and may present familiar propulsion structures. A target that still communicates and controls its attitude gives the servicer a stable object to approach. A silent or tumbling spacecraft requires added sensing, autonomy, fuel, analysis, and retreat capacity. The servicing craft may need to observe the target across several passes before committing to capture.

Access to the target matters just as much as distance. A docking plate allows a planned mechanical connection. A satellite with no servicing interface can force the provider to design around an engine bell, adapter ring, or exposed structure. Solar arrays, antennas, thermal blankets, and plume-sensitive surfaces limit where the rescue craft can move. The Government Accountability Office identified standards, market, and technical barriers affecting in-space servicing in July 2025. A quote should state the assumed target rotation, navigation data quality, capture point, available clearance, approach count, and required final orbit. A change in any one assumption can trigger redesigned tools, added simulations, new flight software, or another test campaign.

Hidden costs

Launch delay is one risk. A rescue vehicle may be complete but unable to fly during the needed orbital window. Storage, battery maintenance, staffing, rocket rebooking, and repeated integration checks can then continue. Another risk appears after launch. Failed checkout, navigation errors, target motion, or a missed capture may extend the operations campaign and consume reserve propellant. Extra tracking support and collision analysis can also enter the bill. A second approach may require new trajectory work rather than a simple repeat of the first command sequence.

The clearest warning comes from programs that create new technology during the rescue project. OSAM-1 began with an estimated range of $626 million to $753 million and later received a $2.05 billion cost commitment, according to program findings reported in 2024. At the upper starting estimate, growth to $2.05 billion equals $1.297 billion, or about 172%. A buyer should ask whether the stated amount includes launch, licensing, insurance, ground stations, post-capture operations, undocking, disposal, schedule reserve, and liability work. A proposal omitting those entries is not an all-in mission figure.

Hidden-cost range Public records do not support one standard add-on percentage. Contingency can run from a limited reserve inside a fixed award to cost growth exceeding $1 billion on a long development program.

Who this cost makes sense for

Rescue spending is easier to defend when the payload still works, the lost function is hard to replace, and the servicing method has been demonstrated. A communications operator can compare the fee with revenue retained during added service years. A science agency can compare it with the expense and delay of building a replacement instrument. Hubble shows the opposite end of the decision. A 2004 GAO review placed a possible shuttle mission at $1.7 billion to $2.4 billion, with gaps in the supporting documentation. Earlier repair history is covered in the account of fixing Hubble’s mirror.

The economic test compares the remaining value of the target with the full rescue bill, replacement cost, launch schedule, and service interruption. A fuel-depleted communications satellite with healthy payloads presents a different case from a science spacecraft with failed power and communications systems. The first may need propulsion support. The second may require several repairs before any mission value returns. Contract reviews should separate the value of restored service from the prestige or technical demonstration attached to the mission.

Makes sense if

• The payload works and added propulsion would restore useful service.
• A replacement would arrive after an unacceptable coverage gap.
• The target has a reachable capture point and manageable rotation.
• Several years of communications, defense, weather, or science output remain.
• The mission uses hardware tested on a similar target.

Doesn’t make sense if

• Power, communications, and payload systems have failed together.
• The object tumbles beyond the rescue craft’s capture limits.
• A replacement is ready at a lower all-in figure.
• The target’s remaining commercial or scientific value is small.
• Launch timing misses the final safe rescue window.

What we verified

• Checked that NASA’s published Swift mission timeline records the $30 million award and compressed development schedule.
• Confirmed that NASA discontinued the OSAM-1 project after an independent review of technical, cost, and schedule problems.
• Cross-referenced the documented MEV docking history for commercial satellite life-extension activity.
• Verified that the published SM3A cost record lists $136 million for Hubble Servicing Mission 3A as historical context.

Article Highlights

• A focused robotic orbit boost has a public contract marker of $30 million.
• Commercial geostationary extension prices are frequently kept private.
• Custom capture, refueling, and robotic assembly can turn a rescue into a multibillion-dollar development program.
• Target rotation, docking access, orbit, launch timing, and remaining payload life drive the purchasing decision.
• Buyers should separate vehicle development, launch, operations, insurance, reserve, relocation, and disposal in the quote.
• Rescue makes financial sense when retained service is worth more than replacement and interruption combined.

Answers to Common Questions

Can a private satellite owner buy a rescue mission?

Yes, but commercial providers focus on spacecraft whose remaining operational value can support a custom aerospace contract. Geostationary communications satellites are established clients because added service years may preserve revenue-producing capacity.

Is refueling cheaper than launching a replacement?

No single public rule applies. Refueling can avoid replacement manufacturing and commissioning, but an unprepared target may require custom robotics and software. The comparison needs the rescue proposal, replacement schedule, remaining payload life, and value of interrupted service.

Does satellite insurance pay for rescue work?

Coverage depends on the policy language and insured event. A policy may address launch loss or in-orbit failure without paying for an elective life-extension mission. The operator must confirm rescue, salvage, liability, and failed-docking treatment with the underwriter.

Can one rescue spacecraft serve several satellites?

Yes. Mission-extension vehicles can detach after completing one assignment and move to another client, provided they retain enough propulsion and the next target is compatible.