Spacecraft navigation has long depended on signals from Earth, whether GPS satellites, ground-based tracking stations, or direct communication with mission controllers. NASA’s Starling mission has now demonstrated something fundamentally different: a system that figures out where it is in space by looking at what surrounds it, including other satellites and orbital debris, and using those objects as reference points. No external signal required.
The System That Reads Its Own Surroundings
The technology behind this capability is called FALCON, which stands for Fast Autonomous Lost-in-space Catalog-based Optical Navigation. It was developed as a joint experiment between NASA and EraDrive, a startup that grew out of Stanford University. The system runs on EraDrive’s Era-Core flight software and uses cameras already present on the Starling spacecraft, specifically the star tracker cameras that satellites typically use to determine their orientation.
What FALCON does is conceptually elegant. It observes objects in the spacecraft’s field of view, identifies them by cross-referencing a catalog of known space objects maintained by the U.S. Department of War, and then uses those confirmed identifications as fixed reference points to calculate the spacecraft’s own orbital position. The logic is similar to how a navigator on a ship might use known landmarks on a coastline to determine their location, except the landmarks here are satellites and pieces of debris moving through low Earth orbit.
This is not a minor refinement of existing methods. It represents the first time a spacecraft has determined its own orbit using optical cameras based on its position relative to other objects in space.
200 Orbits Refined, Autonomously, in Three Days
The Starling demonstration went further than simply proving the navigation concept. A separate set of experiments tested whether the spacecraft could also improve the orbital data it was working with.
Mission controllers loaded a catalog of approximately 20,000 space objects and their predicted orbits onto Starling. FALCON then compared those predictions against its own camera observations. The result: the system’s estimates of object positions turned out to be more precise than the existing catalog data. Over three days, and without any intervention from ground operators, FALCON refined the known orbits of more than 200 objects.
That detail is worth pausing on. Ground-based tracking networks have been the standard source of orbital data for decades. A spacecraft operating autonomously in orbit, using its own sensors and onboard software, produced better positional estimates for more than 200 objects than those networks had on file. Roger Hunter, program manager for NASA’s Small Spacecraft and Distributed Systems program at NASA’s Ames Research Center, described the results as having “far-reaching implications for on-orbit space-traffic monitoring, collision avoidance, and alternative navigation.”
Later this year, the four Starling spacecraft will expand the experiment further. They will share tracking information with one another and use those combined observations to refine their positions collectively, a step toward coordinated autonomous navigation across a network of spacecraft.
Why Autonomous Navigation Changes the Equation
Here is what most coverage of this story tends to underplay: the deeper significance of FALCON is not just about navigation. It is about what becomes possible when spacecraft no longer need to ask permission to know where they are.
GPS works well in low Earth orbit, but its signals weaken with distance and become unreliable or unavailable around the Moon and in deep space. Any mission that ventures beyond GPS range, whether a lunar satellite swarm, a distributed science network, or a crewed mission to Mars, faces a navigation problem that ground-based systems cannot fully solve. Communication delays alone make real-time guidance from Earth impractical at those distances.
FALCON points toward a different model: spacecraft that carry their situational awareness with them. A satellite that can determine its own position, track surrounding objects, and update its own catalog of the space environment does not need to wait for instructions. It can respond to what it observes. For collision avoidance, where timing is critical, that autonomy is not a convenience. It is a safety requirement.
The broader implications extend to how space infrastructure is managed. As the number of objects in orbit grows, the burden on ground-based tracking networks increases. Spacecraft capable of contributing to their own tracking, and to the tracking of objects around them, could help distribute that burden. A fleet of satellites that collectively refines orbital data as it operates is a fundamentally different kind of infrastructure than one that passively receives instructions from the ground.
The FALCON experiment also illustrates how university research transitions into operational technology. The project began as a University SmallSat Technology Partnerships initiative before becoming EraDrive, which is now commercializing its Era-Core software and hardware for broader use.
In Short
NASA’s FALCON system demonstrated that a spacecraft can navigate in orbit without GPS by using other satellites and debris as reference points, a first for optical camera-based autonomous navigation. Over three days, it also refined the orbital data of more than 200 objects without ground intervention, outperforming existing catalog predictions. The significance extends beyond navigation: as missions move farther from Earth, the ability to operate without external signals or real-time ground support becomes essential. FALCON is an early, concrete example of what genuinely autonomous spacecraft infrastructure looks like.
Based on reporting from ScienceDaily AI.