Review of Spacecraft Pose-Estimation Techniques for Close-Proximity Operations

The accurate estimation of spacecraft pose during close-proximity operations is a fundamental capability required for various space missions, including on-orbit servicing, debris management, and formation flying. This review provides a comprehensive overview of the latest techniques employed in spacecraft pose estimation, highlighting both cooperative and non-cooperative scenarios. We discuss a range of algorithmic approaches and sensor technologies, focusing on electro-optic sensors that meet the power and size limitations of modern satellites. The paper further explores the integration of Deep Learning (DL) methodologies, which are gaining traction due to their potential to enhance pose estimation precision. However, significant hurdles remain, including high computational demands and the challenges of transferring learning from synthetic datasets to real-world applications. The paper also emphasizes significant obstacles in determining spacecraft orientation, including variable illumination conditions, computational limitations, and restricted available data. Through an evaluation of current methodologies and the identification of areas necessitating further investigation, this survey aims to advance the development of more robust, efficient, and reliable autonomous navigation systems for future space missions. This research supports SDG 9 (Industry, Innovation, and Infrastructure) by advancing autonomous navigation technologies for safer and more efficient space operations. Enhanced spacecraft pose estimation facilitates responsible space sustainability (SDG 12) through enabling satellite life extension and debris mitigation, and contributes to SDG 13 (Climate Action) by improving precision in satellite-based Earth observation and environmental monitoring.