Starling Formation-Flying Optical Experiment (StarFOX): System Design and Preflight Verification

Abstract

The Starling Formation-Flying Optical Experiment (StarFOX) is intended as the first on-orbit demonstration of autonomous distributed angles-only navigation for spacecraft swarms. StarFOX applies the angles-only Absolute and Relative Trajectory System (ARTMS), a navigation architecture consisting of three innovative algorithms: image processing, which identifies and tracks multiple targets in images from a single camera without a priori relative orbit knowledge; batch orbit determination, which autonomously initializes orbit estimates for visible swarm members; and sequential orbit determination, which continuously refines the swarm state by fusing measurements from multiple observers exchanged over an intersatellite link. Nonlinear dynamics and measurement models provide sufficient observability to estimate absolute orbits, relative orbits, and auxiliary states using only bearing angles without maneuvers. StarFOX will be conducted using a four-CubeSat swarm as part of the NASA Starling mission, and simulations of experiment scenarios demonstrate that ARTMS meets mission performance requirements. Results indicate that mean bearing angle errors are below 35′′ (), initial target range errors are below 20% of true separation, and steady-state range errors are below 2% (). Absolute orbit estimation accuracy is on the order of 100 m. Hardware-in-the-loop tests display robust navigation under a variety of conditions, enabling autonomous, ubiquitous navigation with minimal ground interaction for future distributed missions.