Averaged Spin State Catalog Framework for Uncontrolled Space Objects

We develop a dynamics-based, "shape model free" spin state estimation and prediction framework for uncontrolled space objects. The framework accounts for solar radiation and terrestrial gravity gradient torques, most relevant in medium earth orbit (MEO), geosynchronous earth orbit (GEO), and beyond. Additional averaged perturbations (e.g. eddy current torques) can be incorporated. This framework averages a uniformly rotating object's attitude dynamics over its intrinsic rotation and earth orbit, enabling fast, accurate spin state propagation months or years into the future. By averaging, we distill solar radiation torques to constant curves that are readily modeled and estimated as Fourier series. Requiring no object a priori information and using just light curve-derived synodic spin rate measurements (although other measurement types can be readily incorporated), we estimate an object's inertial spin rate and spin axis as well as its Fourier solar torque and gravity gradient coefficients in a least squares batch filter. Inertial spin axis information is provided by: 1) differences between the inertial and observed (i.e. synodic) spin rate due to time-varying observation geometry, and 2) torque-driven dynamical coupling between the spin rate and spin axis which ties together long-term observations, strongly constraining viable states and breaking solution ambiguities. In GEO simulations with no a priori satellite information, we demonstrate convergence to the true state within one year using sparse synodic spin rate measurements. We apply the framework to real light curves of the defunct GEO satellite Telstar 401 and compare our spin pole estimate to independent Deep Space Network radar solutions, showing clear consistency. This light-weight framework can be readily applied to existing light curve databases, some of which span many years for specific objects. State estimates can be used for long-term spin state prediction and updated through catalog maintenance observations. (c) 2026 The Author(s). Published by Elsevier B.V. on behalf of COSPAR. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).