Abstract—This paper presents a distributed framework for angleonly relative orbit determination in satellite networks, motivated by the increasing need for autonomous and scalable navigation in congested orbital environments. Angles-only approaches are getting a lot of attention because they rely on passive optical sensors, which are lightweight, low-power, and already available on most spacecraft. However, single-observer configurations are strongly limited by poor observability, leading to fragile and inaccurate estimates. To overcome these challenges, we introduce a complete pipeline that integrates three key components: (1) a geometric 2026 IEEE Aerospace Conference | 979-8-3315-7360-7/26/$31.00 ©2026 IEEE | DOI: 10.1109/AERO66936.2026.11520155 initialization algorithm that estimates the initial target state from synchronized line-of-sight measurements of multiple observers; (2) an analytical covariance derivation that quantifies the accuracy of the initial estimation, independently of orbital dynamics; and (3) a distributed extended Kalman filter (DEKF) based on information consensus, that fuses measurements across the network to provide continuous tracking of the target spacecraft. Numerical simulations in a realistic low Earth orbit scenario demonstrate that the proposed method achieves estimation performance close to a centralized EKF, while maintaining full distribution and robustness to network topology. The DEKF ensures consensus among observers and limits position errors to a few meters and velocity errors below 10−4 km/s, significantly outperforming independent filters. These results prove that consensus-based distributed angle-only navigation can provide accurate and scalable orbit determination capabilities, paving the way for future multi-satellite missions and space-based surveillance architectures.
Distributed Angle-Only Relative Orbit Determination in Satellite Networks
Romagnoli G.;Bucchioni G.;Falcone G.
2026-01-01
Abstract
Abstract—This paper presents a distributed framework for angleonly relative orbit determination in satellite networks, motivated by the increasing need for autonomous and scalable navigation in congested orbital environments. Angles-only approaches are getting a lot of attention because they rely on passive optical sensors, which are lightweight, low-power, and already available on most spacecraft. However, single-observer configurations are strongly limited by poor observability, leading to fragile and inaccurate estimates. To overcome these challenges, we introduce a complete pipeline that integrates three key components: (1) a geometric 2026 IEEE Aerospace Conference | 979-8-3315-7360-7/26/$31.00 ©2026 IEEE | DOI: 10.1109/AERO66936.2026.11520155 initialization algorithm that estimates the initial target state from synchronized line-of-sight measurements of multiple observers; (2) an analytical covariance derivation that quantifies the accuracy of the initial estimation, independently of orbital dynamics; and (3) a distributed extended Kalman filter (DEKF) based on information consensus, that fuses measurements across the network to provide continuous tracking of the target spacecraft. Numerical simulations in a realistic low Earth orbit scenario demonstrate that the proposed method achieves estimation performance close to a centralized EKF, while maintaining full distribution and robustness to network topology. The DEKF ensures consensus among observers and limits position errors to a few meters and velocity errors below 10−4 km/s, significantly outperforming independent filters. These results prove that consensus-based distributed angle-only navigation can provide accurate and scalable orbit determination capabilities, paving the way for future multi-satellite missions and space-based surveillance architectures.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
