A recent study introduces advanced techniques to enhance orbit determination (OD) for large constellations of Low Earth Orbit (LEO) satellites. By leveraging Global Navigation Satellite System (GNSS) observations and inter-satellite ranging, these methods significantly improve both accuracy and computational efficiency—key factors for satellite communication, remote sensing, and navigation augmentation.
RMSE of the stepwise and centralized dynamic OD. (a) RMSE of the Stepwise OD (b) RMSE of centralized OD. Image Credit: Satellite Navigation
LEO satellite constellations play a crucial role in modern space-based applications, from global connectivity to Earth observation. However, tracking these satellites presents a major challenge due to their sheer numbers and the need for continuous high-precision data. While ground-based tracking stations have limitations in managing large constellations, spaceborne GNSS receivers offer a promising alternative. Yet, current approaches still struggle with balancing accuracy and computational demands, highlighting the need for more sophisticated solutions.
A study published on February 10th, 2025, in Satellite Navigation by researchers from the Xi'an Research Institute of Surveying and Mapping and the State Key Laboratory of Spatial Datum introduces a stepwise autonomous OD method for large LEO constellations. By integrating GNSS data with inter-satellite ranging, the study offers a more precise and computationally efficient approach to orbit tracking.
The research presents three innovative autonomous OD strategies. The first method fuses GNSS data with inter-satellite link (ISL) range measurements to refine orbit parameters. The second uses ISL ranges as constraints, improving accuracy without increasing computational complexity. The third approach dynamically adjusts the covariance matrix of orbit predictions, compensating for errors caused by anomalies in dynamic models.
These methods start with an initial orbit parameter estimation using GNSS observations, followed by refinements with ISL range data. The adaptive strategy is particularly noteworthy, as it fine-tunes the covariance matrix with an adaptive factor to control dynamic model errors. Simulations demonstrate substantial improvements, with the root mean square error (RMSE) of position estimates dropping to as low as 11.34 cm when dynamic models are combined with ISL ranges. Additionally, the ability to parallelize orbit estimations for individual satellites reduces computational load, making this approach highly scalable for large constellations.
Our stepwise autonomous OD methods provide a practical solution to the computational and accuracy challenges faced by large LEO constellations. By integrating GNSS observations and ISL ranging, we achieve higher precision and efficiency, paving the way for more robust satellite operations.
Dr. Yuanxi Yang, Study Author, Chinese Academy of Sciences
The impact of this research extends far beyond orbit determination. These enhanced OD techniques provide a scalable framework for improving satellite communication, remote sensing, and navigation augmentation. As LEO constellations continue to expand in both size and complexity, these methods offer a reliable foundation for maintaining precise orbit control—unlocking new possibilities in global navigation, environmental monitoring, and beyond.
Journal Reference:
Yang, Y., & Song, X., et al. (2025) Stepwise autonomous orbit determination of large LEO constellations by GNSS observations with partial inter-satellite ranging. Satellite Navigation. doi.org/10.1186/s43020-025-00160-1