Relative Navigation Architecture for Multistatic SAR Formations

Abstract

Synthetic Aperture Radar (SAR) is widely recognized as a valuable tool for Earth Observation and environmental monitoring due to its ability to operate effectively under all weather and lighting conditions. However, traditional SAR systems face inherent limitations in simultaneously achieving wide coverage and high spatial resolution. Distributed Synthetic Aperture Radar (DSAR) systems overcome these constraints by using multiple coordinated receivers and transmitters in conjunction with Digital Beamforming (DBF) techniques, thus enabling High-Resolution Wide-Swath (HRWS) imaging. Despite these advantages, successful implementation of DSAR critically depends on accurate control of the relative positions and orientations of satellite platforms, which relies on precise relative navigation information. Current navigation solutions primarily depend on Global Navigation Satellite Systems (GNSS), but this introduces vulnerabilities such as signal disruptions, jamming, and spoofing. This thesis addresses these limitations by proposing an autonomous, robust, real-time relative navigation architecture specifically designed for small-platform DSAR formations. The architecture integrates multiple sensing modalities-including GNSS, Vision-Based Sensors (VBS), and Radio-Frequency (RF) sensors-to enhance redundancy and resilience against GNSS disruptions. Additionally, a dedicated simulation tool was developed to assess the accuracy and effectiveness of the proposed multi-sensor navigation system under various configurations, supporting robust and continuous baseline estimation essential for successful DBF operations. The proposed architecture requires a total power of 10 W and a mass of 1.5 kg for each satellite in the formation. This work analyzes two distinct formation strategies: an Along-Track (AT) Formation, characterized by satellites that, without continuous active control, drift apart due to relative orbital dynamics, thus requiring strict control of their relative motion; and a Passive Safety Ellipse (PSE) Formation, designed to maintain relative trajectories within a predefined safety tube radius, ensuring passive collision avoidance. Results from simulations show that the integrated navigation system achieves a baseline estimation accuracy in terms of 3D Root Sum Square (RSS) error of 0.5 cm for the uncontrolled AT formation, 2 cm for the controlled AT formation, and approximately 1 cm for the PSE formation, demonstrating the efficacy of the proposed multi-sensor navigation approach