A 22-year-old telescope is sliding out of the sky, and NASA has decided it is worth $30 million to catch it. The rescue flies Tuesday, June 30.
That morning, at 6:23 a.m. EDT (1023 GMT), the last Pegasus XL rocket ever built is scheduled to drop from beneath a carrier plane over the Reagan Missile Test Site on Kwajalein Atoll, in the Marshall Islands, and light its motors. Tucked inside is a spacecraft the size of a refrigerator, named Link, with one job: find the Neil Gehrels Swift Observatory, grab hold of a satellite that was never designed to be grabbed, and shove it back to a safer altitude before Earth’s atmosphere finishes pulling it down.
Nobody has done this before. A privately built robot has never captured and reboosted a working U.S. government science satellite that lacks any docking port, handle or grappling fixture. If it works, Swift buys at least five more years of life. If it fails, the telescope falls anyway — which is exactly what it is going to do if no one tries.
Why a falling telescope is worth saving
Swift launched in 2004 on a two-year mission and never stopped working. Its trick is speed. Somewhere in the cosmos a massive star explodes roughly every second, and Swift was built to whip around and stare at the resulting gamma-ray bursts — the most violent explosions since the Big Bang — within minutes of detecting them. The Hubble Space Telescope, by comparison, can take up to two days to swing onto a new target.
“This is an observatory with unique capabilities for astrophysics. … It is a Swift observatory that can quickly pivot across the night sky to find things that go boom in the night.”
Shawn Domagal-Goldman, director of NASA’s Astrophysics Division, in a June 17 briefing
The name, it turns out, is not an acronym. “It comes from the ability to rapidly and autonomously repoint its narrowfield X-ray and UV telescopes almost anywhere on the sky,” said Brad Cenko, Swift’s principal investigator, who works out of Penn State. The telescope earned its keep: its data helped confirm that gold, platinum and the universe’s heaviest elements are forged in those bursts, and in 2022 it caught a blast so bright astronomers nicknamed it the BOAT, for Brightest of All Time.
The problem is altitude. Swift carries no engine. It started in an orbit 375 miles (600 kilometers) up, but atmospheric drag has been dragging it down for two decades, and a stretch of heightened solar activity has puffed up the upper atmosphere and accelerated the fall. The observatory is now on track to drop below 186 miles (300 km) by October — low enough that Link might not be able to reach it in time. NASA parked the telescope in a low-power mode in February, halting science to slow the descent and buy a few weeks. It is the same physics of orbital decay that keeps shrinking the margins for everything flying in low Earth orbit.
A spacecraft built in nine months
NASA picked Katalyst Space Technologies, a young company in Flagstaff, Arizona, only nine months ago, in September 2025. The firm had been planning a demonstration flight; when NASA raised the alarm about Swift, it pivoted to a live rescue. Going from a blank sheet of paper to a finished spacecraft bolted to a rocket in under a year is not how space hardware usually gets built.
Link is small but busy — three ion engines, three robotic arms, and a suite of sensors and thrusters. After launch it will spend several weeks in orbit running checkouts, then attempt the rendezvous, latch onto a feature of Swift’s body with its arms, and spend up to three months gently nudging the telescope higher.
“There’s a lot of very simple things that can go wrong, and we’re adding a lot of additional complexities to the program, but we’ve been through an aggressive test campaign over the last few months.”
Kieran Wilson, Link’s principal investigator at Katalyst
The urgency is the whole point of the design, Wilson said: every shortcut and every all-nighter traces back to the deadline Swift’s orbit imposed. That speed-first model is the same bet driving the rest of the commercial space sector’s scramble — build fast, fly fast, prove the capability before a competitor does.
The bigger prize behind one telescope
Reboosting Swift is not really about one telescope. NASA looked at hiring industry to lift the aging Hubble in 2022 — a proposal from entrepreneur Jared Isaacman and SpaceX — but backed off, unwilling to risk an instrument that should keep working into the mid-2030s. Swift is different: it is failing now, with no replacement waiting. That makes it a low-stakes place to try a high-stakes maneuver.
The capability also has a harder edge. China demonstrated on-orbit servicing in 2022, using its SJ-21 spacecraft to tow a defunct satellite, and the Pentagon wants American craft that can move and rescue assets in orbit. Katalyst already holds a U.S. Space Force contract to demonstrate something similar with a larger vehicle, Nexus, due to fly in 2027. A clean grab of Swift would be the proof of concept.
Cenko, the scientist with the most to lose, has made his peace with the gamble. He admits to sleepless nights. He also does the math out loud: the alternative is watching Swift burn up for nothing.
“Maybe we can’t be doing science for a few more months. But trading that against the potential benefit of many years of upsurge? That is a no-brainer.”
Brad Cenko, Swift principal investigator, speaking to Space.com
For now the whole bet rides on a rocket that won’t fly again after this, a robot that has never done this, and a telescope that can only wait. Tuesday morning, over a speck of coral in the Pacific, they find out.
