SpaceX Demands Orbital Coordination After Starlink Satellites Pass Within Tens of Meters of Other Spacecraft
SpaceX has publicly called for better coordination in low Earth orbit after a series of close approaches between Starlink satellites and other spacecraft. According to the company, the separation distances ranged from tens to hundreds of meters, which it called "too close for comfort." In orbital mechanics, that is very little: trajectory prediction errors are typically comparable to such distances. Starlink remains the largest constellation in orbit, so every poorly coordinated launch by a competitor becomes its problem. • Close approaches: from tens to hundreds of meters. • Reason for the complaints: not enough data sharing between operators. • Key question: whether common rules for launches and maneuvers are needed.
AI-processed from Ars Technica; edited by Hamidun News
SpaceX says orbit needs tighter coordination between operators: according to the company, recent close approaches between Starlink satellites and other spacecraft ended at distances of tens to hundreds of meters. The quote from its statement is short and blunt: "Too close for comfort." In near-Earth space, where speeds reach about 7.5 km/s, a distance like that translates into a time difference of mere fractions of a second.
The available excerpt of the Ars Technica material does not disclose all the details, including who was involved in the specific incidents. What follows therefore relies on what was said and on public context.
Why "hundreds of meters" is a lot of trouble
On paper, hundreds of meters look safe. In practice, they are not. An object's predicted position is built from observation data and models, and the error a day ahead can easily reach hundreds of meters, or more for spacecraft that do not share ephemerides. If an operator does not share precise trajectories, the other party has to rely on radar observation catalogs. Those are updated less often and are less accurate.
SpaceX has its own stake here. Starlink already numbers thousands of satellites at altitudes of around 500–550 km, and the company has published collision-avoidance maneuver reports for years. It sets its own threshold stricter than the industry's: it reacts at a collision probability on the order of 1 in 100,000, while the commonly used cutoff is 1 in 10,000. The denser the constellation, the more false alarms and unnecessary maneuvers it faces if a neighbor stays silent about its plans.
What has already happened
The complaints are not new. In December 2025, SpaceX representatives already complained about the launch of a Chinese rocket whose spacecraft, by their account, passed about 200 meters from a Starlink satellite without prior coordination. The new statement continues the same line: the company wants to turn private complaints into a public conversation about rules.
Key figures from the statement and context:
- close-approach distance: from tens to hundreds of meters;
- Starlink operating altitudes: around 500–550 km;
- orbital speed: about 7.5 km/s;
- SpaceX maneuver threshold: about 1 in 100,000 collision probability;
- main shortfall: real-time sharing of precise ephemerides.
What operators and regulators should do
The practical conclusion is simple: without a common data protocol, everyone suffers. Operators should publish planned trajectories and launch windows in advance, not after orbital insertion. Launch companies need pre-launch conjunction screening that accounts for constellations already in flight, not only large cataloged objects.
For regulators, the question is harder. Voluntary agreements currently work only where the parties are willing to talk. If SpaceX manages to win even a single data-exchange format and minimum notification periods, that would reduce the number of emergency maneuvers and the amount of satellite propellant they consume. Those building their own constellations would do well to build data sharing into the architecture from the start: retrofitting a system after an incident costs more and takes longer.
With no new details available yet, the main signal lies elsewhere: the largest private constellation in orbit is publicly admitting that its own measures are no longer enough.
Key Questions Answered
Why "hundreds of meters" is a lot of trouble?
On paper, hundreds of meters look safe. In practice, they are not. An object's predicted position is built from observation data and models, and the error a day ahead can easily reach hundreds of meters, or more for spacecraft that do not share ephemerides. If an operator does not share precise trajectories, the other party has to rely on radar observation catalogs. Those are updated less often and are less accurate. SpaceX has its own stake here. Starlink already numbers thousands of satellites at altitudes of around 500–550 km, and the company has published collision-avoi
What has already happened?
The complaints are not new. In December 2025, SpaceX representatives already complained about the launch of a Chinese rocket whose spacecraft, by their account, passed about 200 meters from a Starlink satellite without prior coordination. The new statement continues the same line: the company wants to turn private complaints into a public conversation about rules. Key figures from the statement and context: close-approach distance: from tens to hundreds of meters; Starlink operating altitudes: around 500–550 km; orbital speed: about 7.5 km/s; SpaceX maneuver threshold: abou
What operators and regulators should do?
The practical conclusion is simple: without a common data protocol, everyone suffers. Operators should publish planned trajectories and launch windows in advance, not after orbital insertion. Launch companies need pre-launch conjunction screening that accounts for constellations already in flight, not only large cataloged objects. For regulators, the question is harder. Voluntary agreements currently work only where the parties are willing to talk. If SpaceX manages to win even a single data-exchange format and minimum notification periods, that would reduce the number of e
Delegate Deep Research, Coding & Workflows to an AI Agent
Deploy autonomous multi-step execution with NomadNet Agent (agent.nomadnet.ai) — automated web research, data synthesis, code generation, and complex task orchestration.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.