NASA's ACS3 mission aims to be the first Earth-bound solar sail to execute calibration steering laws for in-orbit estimation of solar-sail acceleration parameters. To maximise the mission's scientific return, this study identifies the physical effects to include in the dynamical model, the solar-sail acceleration parameters observable from flight data, and the uncertainties to consider during the orbit determination process. The sensitivity of the solar-sail dynamics to perturbations, model uncertainties, and sail-attitude errors is investigated by 1) comparing a reference orbit with modified orbits, each altered in a single dynamical aspect, and 2) evaluating the accuracy of modified models in reconstructing the reference orbit through iterative initial state adjustments. For the one-sigma 10-meter observation noise level of the ACS3 mission and a seven-day arc, results indicate that higher-order lunar perturbations, planetary third-body effects, and relativistic corrections can be omitted from the dynamical model. Additionally, the geopotential expansion may be limited to degree and order 32. In contrast, the dynamics should include the effects of solid Earth tides, account for the instantaneous Sun-sailcraft distance in the solar radiation pressure model, and assume imperfect reflection from the sail surface in the solar and planetary radiation pressure models. Furthermore, the analysis reveals varying levels of observability for the sail optical coefficients, with frontside reflectivity and specularity showing the strongest influence on the solar-sail dynamics. Finally, systematic attitude errors and uncertainties in atmospheric density and accommodation coefficients are the most challenging factors to absorb through initial state adjustment, potentially complicating the estimation of solar-sail acceleration parameters.

Solar sailing exploits solar radiation pressure to generate propellantless thrust, enabling mission applications beyond the capabilities of conventional propulsion systems. Despite this potential, the lack of in-flight validation for solar-sail force models has limited confidence in applying solar sailing beyond technology demonstration missions. This study presents the first comprehensive investigation into the potential of solar-sail performance characterisation from flight data by applying a covariance-based estimation framework using simulated GNSS observations for NASA’s ACS3 mission.A set of calibration steering laws is proposed to facilitate the in-orbit estimation of the parameters governing the solar-sail acceleration. The study focuses on the sail frontside reflectivity and specularity, the optical coefficients exerting the strongest influence on the solar-sail dynamics. For each steering law, the covariance analysis quantifies the achievable estimation accuracy of these coefficients as a function of measurement noise, observation arc length, sampling rate, and ACS3 expected orbital evolution over the coming year. The operational feasibility of the calibration steering laws is also assessed through the evaluation of power budget, ground station communication, altitude maintenance, sail material degradation, and attitude rate limitations.For the 10-meter observation noise level expected in ACS3 telemetry, results indicate that a dedicated in-flight calibration can reduce the formal errors of the optical coefficients in the (Formula presented) to (Formula presented) range, an improvement of two to three orders of magnitude compared to pre-flight ground characterisation. When estimation performance is evaluated against operational constraints, the power budget is identified as the main limiting factor, and the fixed in-plane pointing steering law emerges as the most robust strategy, consistently delivering high-accuracy estimates while satisfying all operational constraints across diverse orbital geometries.