NASA Calls Off Swift Rescue After LINK Control Problems Persist

Katalyst Space Technologies regained partial control of its spinning LINK spacecraft, but not enough to complete the job that mattered most. NASA and Katalyst have called off the planned capture and orbit raise of NASA’s Neil Gehrels Swift Observatory because of an ongoing attitude-control issue.

LINK may still approach Swift to demonstrate rendezvous and proximity operations. It will not grapple the observatory or move it into a higher orbit, however. NASA anticipates that Swift, which has operated since 2004, is likely to re-enter Earth’s atmosphere later in 2026 without intervention. The outcome turns a rescue mission into a narrower technology demonstration and illustrates why orbital servicing depends on control authority before robotic arms can provide any useful actuation.

Recovery did not restore the complete mission

LINK entered a multi-axis spin of about 9 degrees per second, producing sporadic communications and preventing progress toward Swift. Engineers used one of the spacecraft’s electric propulsion thrusters to slow that rotation to about 1.47 degrees per second. Katalyst then uploaded flight software with new attitude-control algorithms intended to stabilize LINK using its remaining actuators.

That was a meaningful recovery, but reducing a spin is not the same as establishing the pointing precision and control margin needed for capture. A servicing spacecraft must navigate near another vehicle, match its relative motion and operate its robotic hardware without allowing those motions to destabilize the combined system. NASA and Katalyst have not detailed the final technical assessment behind their decision, beyond citing the continuing attitude-control issue.

The distinction matters because robotic servicing joins several tightly coupled systems. Navigation sensors must estimate relative position and motion; guidance software must plan a safe approach; actuators must control the servicer’s orientation; and robotic arms must capture a client that may never have been designed for servicing. Once contact occurs, the mass, dimensions and dynamic response of the system change. An actuator shortfall that can be managed during free flight may therefore remain unacceptable for grappling and orbit raising.

NASA awarded Katalyst the Swift mission in September 2025, and LINK launched July 3, 2026. The unusually compressed schedule was driven by Swift’s decaying low Earth orbit. Increased solar activity raised atmospheric drag, accelerating that decline. NASA describes the project as a high-risk attempt whose remaining rendezvous work could still produce operational data for later servicing missions, but it no longer presents a pathway to extending Swift’s life.

Optus D3 follows a different servicing model

A separate Northrop Grumman mission shows another route to satellite life extension. Its Mission Robotic Vehicle and associated propulsion pods are traveling toward geostationary orbit, where the vehicle is planned to use two robotic arms to install a Mission Extension Pod on the nearly 17-year-old Optus D3 communications satellite.

The pod will augment D3’s propulsion, helping it maintain its orbital position. Optus says installation is scheduled for late 2027 and could extend the satellite’s operating life by up to six years. That remains a planned result, not a completed upgrade. The project has nevertheless passed an earlier operational milestone: a Mission Extension Vehicle provided propulsion and attitude-control support to D3 for more than a year before successfully undocking in the first phase of Project Aurora.

The Northrop Grumman architecture separates the reusable servicer from the propulsion unit left with the client. The robotic vehicle can carry a pod to a satellite, install it and later move to other assignments. That potentially spreads the cost and utility of a sophisticated sensing, navigation and manipulation platform across multiple service calls, rather than dedicating the entire vehicle to one satellite.

It also creates demanding integration work. Arm movement produces reaction forces that the spacecraft control system must counter. After attachment, the pod must become a reliable part of a satellite that was already operating in orbit. Ground testing can validate hardware and software, but it cannot reproduce every aspect of free-flight dynamics and contact in microgravity.

Both missions pursue the same broad economic objective: preserve a useful spacecraft without launching a complete replacement or sending astronauts to perform repairs. No verified savings have yet been established for these specific operations, and the technical payoff depends on successful rendezvous, capture, installation and long-term operation.

LINK demonstrates the boundary most clearly. Engineers recovered communications and sharply reduced an uncontrolled spin, yet the remaining control problem was enough to cancel Swift’s capture and boost. Its next test is no longer whether it can save the observatory, but how much reliable rendezvous data it can gather before Swift’s expected re-entry later this year.

By Jonathan Barrett — Editor for AMI’s future mobility and autonomous systems section, with two decades covering robotics, e-mobility, drone-vehicle convergence, and transport mechanical systems.