Katalyst’s LINK reenters after its attempt to rescue NASA’s Swift telescope

The story

Katalyst Space’s LINK spacecraft has reentered Earth’s atmosphere, closing an ambitious commercial attempt to reach, capture and raise the orbit of NASA’s Neil Gehrels Swift Observatory. The company said LINK came down on September 25 after 85 days in orbit. It never completed the planned grapple or boost, leaving Swift on course for its own eventual reentry, but the mission provides a rare public record of how quickly a novel servicing system can be fielded—and how one hardware failure can reshape an entire orbital campaign.

NASA awarded Katalyst the $30 million contract in September 2025. The company then designed, built, tested and launched LINK in nine months, sending it to orbit aboard Northrop Grumman’s Pegasus XL rocket on July 3, 2026. Its target was especially difficult: Swift, launched in 2004 to study gamma-ray bursts and other fast-changing cosmic events, was not designed for servicing and has no standardized docking interface. LINK therefore carried three robotic arms intended to grip the larger observatory and move it to a more durable orbit.

The spacecraft initially deployed its solar arrays, established communications and began commissioning its avionics, propulsion, rendezvous sensors and robotics. During that process, however, an anomaly affected attitude control and sent LINK into a multi-axis spin. Katalyst said an electrical power-system fault made two of the spacecraft’s three reaction wheels unavailable. Reaction wheels normally let a spacecraft change orientation without continuously burning propellant, so losing two sharply reduced the available control options.

Engineers restored communications and developed new guidance, navigation and control modes that combined small reaction-control thrusters with the remaining wheel. Hall-effect thrusters reduced the body rate from nine degrees per second to 1.47 degrees per second, allowing the team to upload revised flight software and recover limited control. That was a meaningful on-orbit engineering achievement, but it carried a decisive cost: using chemical thrusters to manage attitude consumed propellant much faster than planned.

Katalyst and NASA concluded that LINK no longer had enough fuel to execute the capture and orbit-raising sequence safely. The mission was redirected toward proximity operations and technical tests. LINK eventually approached within roughly 12 to 15 kilometres of Swift, collected unresolved imagery, exercised all three robotic arms and practised grapple motions with its grippers. Those demonstrations did not rescue the observatory, and they should not be described as validation of a complete servicing mission. They did generate flight data across systems that had previously been tested mainly on the ground.

Swift’s scientific value explains why NASA accepted the risk. Designed for a two-year mission, it has operated for more than two decades and helped astronomers respond rapidly to some of the universe’s most energetic events. Its orbit has been falling faster as the upper atmosphere expands during heightened solar activity, increasing drag. NASA resumed science observations with two instruments after the boost plan was abandoned, but the agency has said the observatory is expected to reenter, ending a remarkably productive mission.

The episode matters beyond one telescope. Governments and satellite operators increasingly want spacecraft that can inspect, refuel, repair, reposition or retire other assets in orbit. Those capabilities could extend mission life, reduce replacement costs and help manage congestion. Yet servicing an object that was never built to dock is fundamentally different from visiting a cooperative target with standardized fixtures. LINK’s experience shows that robotic arms are only one part of the system: power distribution, attitude control, software recovery, propellant margins and operational decision-making are equally mission-critical.

Katalyst says the lessons will feed into NEXUS, its planned multi-mission robotic platform, and that four follow-on missions or programs are in development. The company highlighted a Defense Innovation Unit deorbiting program and an on-orbit power-beaming project with the U.S. Naval Research Laboratory. These plans remain future work. The relevant measure will be whether the company converts LINK’s anomaly data into design changes that can be demonstrated on subsequent flights, rather than treating schedule speed alone as proof of readiness.

INNOVOX analysis: LINK is best understood neither as a successful rescue nor as a useless failure. Its central objective was not achieved, yet the mission exposed failure modes that simulations and ground tests did not remove. For an emerging orbital-servicing industry, that evidence can be valuable—but only if operators disclose root causes, quantify what was demonstrated and verify corrective actions. The strongest innovation story here is institutional: NASA accepted a compressed, high-risk commercial mission to preserve a scientific asset while generating capabilities that may serve later missions.

What to watch next is the technical follow-through. Katalyst has not publicly completed a detailed root-cause account of the power-system fault or explained the design changes NEXUS will inherit. Future launches should show greater reaction-wheel and power redundancy, realistic propellant reserves for anomaly recovery, and increasingly difficult rendezvous tests before attempting another non-cooperative capture. Swift’s remaining orbit and controlled shutdown will also require close attention. The long-term verdict on LINK will depend less on the rhetoric surrounding its reentry than on whether its data make the next servicing spacecraft measurably safer and more capable.

INNOVOX analysis

LINK shows why on-orbit servicing cannot be judged by robotic hardware alone. Power redundancy, attitude control, recovery software and propellant margins determine whether a spacecraft can turn proximity into useful work. The mission’s value now depends on transparent root-cause analysis and verified improvements on later flights.

What to watch

Watch for a detailed account of the electrical fault, published corrective actions for NEXUS, stronger subsystem redundancy, follow-on rendezvous demonstrations and NASA’s management of Swift as the observatory’s orbit continues to decay.