On March 16, Sustain Space launched a satellite called Xiyuan-0 on a Kuaizhou-11 rocket from Jiuquan Satellite Launch Center and spent the following weeks testing a flexible robotic arm through four operating modes. The results, published in more technical detail than most Chinese commercial space companies release, suggest that on-orbit satellite refueling is less a theoretical exercise and more an impending infrastructure question — one that China's state-owned enterprises and private companies are now chasing in parallel.
Sustain Space, incorporated in June 2022 as a commercial subsidiary of Emposat, verified four modes during its demonstration: a pre-programmed autonomous refueling simulation, human teleoperation from the ground, vision-based servo operations, and force-controlled drawing tests. All four returned green, according to SpaceNews. The arm uses a hollow continuum design — spring-like tubes connected by motors that pull internal cables, with electrical and eventually propellant lines routed through the structure rather than across the satellite exterior. That is a meaningful choice for refueling work: exposed lines snag on anything that moves.
Wang Xueqian, a researcher at Tsinghua University's Shenzhen International Graduate School who has spent over a decade on space robotics, led the development team, which also drew on collaborators at Shenzhen Mofang and Hunan University, according to Xinhua. Sustain Space has been clear about its business focus: satellite life extension and orbital debris management. It is not a stealth startup, and it is also not a company with the budget of a major aerospace prime.
"This is quite a technological accomplishment. I am even more impressed with the level of detail in the company statement," said Victoria Samson, a space sustainability researcher at the Secure World Foundation, in a published response to the announcement. Samson has reviewed hundreds of on-orbit servicing proposals. Her baseline for novelty is calibrated.
China's state-owned enterprises have been at this longer. The Shanghai Academy of Spaceflight Technology, a state-owned enterprise, conducted what it described as a world-leading geostationary refueling test between the Shijian-25 and Shijian-21 satellites in 2025, topping up storeable propellants on Shijian-21, according to China-in-Space. That demonstrated that GEO refueling is physically possible at operational altitude — roughly 35,786 kilometers, where most high-value communications satellites live. Sustain Space's contribution is the same trick executed by a commercial company, with a different mechanical architecture. The Xiyuan-0 arm is the first flexible robotic arm for in-orbit refueling developed outside a Chinese state-owned enterprise, per China-in-Space.
The commercial case is not hypothetical. Between 10 and 20 geostationary satellites reach end-of-life every year because they run out of fuel, according to US Air and Space Forces data, even though their electronics, antennas, and structures remain functional. A communications satellite in GEO that costs $400 million to build and launch is a candidate for refueling if the fuel depletes in year 15 of a 20-year design life. That arithmetic is why the US Space Force has four GEO satellite servicing or refueling missions planned for 2026: the Astroscale Refueler, SpaceLogistics' Mission Extension Vehicle with its Robotic Servicing Gateway System, Tetra-5, and Kamino. NASA tried to own this capability with a project called OSAM-1 — on-orbit servicing, assembly, and manufacturing — and cancelled it in March 2024 after nearly a decade of work, citing technical, cost, and schedule challenges, per NASA. The US approach is now firmly commercial, and the missions are imminent.
What Sustain Space demonstrated on Xiyuan-0 does not yet mean satellite life extension is a working business. The arm operated on a test satellite, not a live client vehicle. Autonomous refueling simulation is not the same as connecting to a real fuel port on a customer spacecraft, which requires rendezvous, proximity operations, and the kind of cooperative docking that SpaceLogistics and Astroscale have been building toward for years. The arm's flexibility is also a design choice with tradeoffs: a continuum arm bends around obstacles but typically generates less precise endpoint positioning than a rigid-link arm. Whether that trade is right for operational refueling is a question the demonstration does not answer.
Xiyuan-0 carries a second technical demonstration — an inflatable drag sphere that will expand to 2.5 meters in diameter at the end of the test campaign, deliberately increasing atmospheric drag to accelerate deorbit from decades to under one year, according to NASA Spaceflight. That is a responsible way to clear a test article from GEO. It is also a reminder that the satellite was not designed to stay.
The Sustain Space demonstration is real. The four modes worked. The technical detail in the company's public statement is unusually thorough for a Chinese commercial space company, which is itself worth noting — either they are building toward a real service or running a sophisticated marketing operation, and the data they released is more consistent with the former. What remains unproven is whether a hollow continuum arm is the right architecture for operational refueling, whether the economics close for commercial customers, and whether Sustain Space can land a first contract. Those are questions for 2027 and beyond.
The broader signal is clearer: on-orbit refueling is no longer a research topic. It is a contracting discussion.