Maintaining the stringent orbital requirements of Copernicus Earth-observation missions, such as Sentinel-1 and Sentinel-2, entails significant manual effort from operations teams. These missions, funded by the EU and ESA, operate in sun-synchronous frozen orbits with repeating ground tracks, where payload performance depends on maintaining both the ground track close to its reference and a near-frozen eccentricity. Operational tasks, including ground-track control, tandem-phase formation keeping, and orbital repositioning, therefore require repeated optimisation of in-plane (IP) and out-of-plane (OOP) manoeuvres under the effects of multiple orbital perturbations.
This paper presents further improvements to an enhanced semi-analytical optimisation algorithm developed to reduce this workload while providing accurate manoeuvre planning across multiple Copernicus missions and mission phases, and potentially soon for reference orbital acquisition operations during the LEOP phase. The upgraded Earth Observation Orbit Control (EOOC) algorithm incorporates a refined dynamical model that extends beyond the circular-orbit assumption. The main perturbations driving the long-term behaviour of low-Earth sun-synchronous orbits, namely the J? effect and atmospheric drag, are included, and solar radiation pressure contributes to the eccentricity dynamics. Assuming a constant drag coefficient, the method propagates orbital deviations over several months and optimises complete chains of manoeuvres rather than single events.
The algorithm uses updated semi-analytical equations to compute manoeuvres directly from J?-driven orbital evolution and incorporates real-time mass corrections, enabling manoeuvre durations to match operational expectations. Additionally, the spacecraft yaw, pitch, and roll angles are integrated into the computation of the thrust direction, further increasing prediction accuracy. Together, these upgrades yield significantly more accurate long-term trajectory predictions.
The algorithm has demonstrated robust performance across diverse operational scenarios, supporting routine operations for two years on Copernicus Sentinel-1A, and more recently on Copernicus Sentinel-1C/D and Copernicus Sentinel-2A/B/C, generating efficient automated manoeuvre plans with far less manual intervention than previously required by operations teams. Its extended prediction horizon, unified IP/OOP optimisation, and improved modelling of thrust geometry have greatly enhanced planning efficiency and manoeuvre consistency. Moreover, this software will be used for the tandem phase of Copernicus Sentinel-2A/B in February, 2026, and for the reference-orbit acquisition of Copernicus Sentinel-3C at the end of 2026.
Conclusion: The enhanced semi-analytical approach provides a versatile, scalable, and operationally reliable tool for orbit maintenance across current and future Copernicus missions. By substantially reducing manual optimisation workload and delivering reliable long-term manoeuvre plans with higher fidelity, it strengthens mission operations and ensures continued compliance with orbital constraints essential for high-quality Earth-observation data.