GNSS-Driven Closed-Loop Navigation and Attitude Control Architecture for Small Satellites

Global Navigation Satellite Systems (GNSSs) are widely used for spacecraft orbit determination, whereas attitude estimation and control typically rely on inertial or optical sensors. This work presents a GNSS-driven closed-loop navigation and attitude control architecture in which spacecraft position, velocity, attitude, angular rate, and receiver clock bias are jointly estimated using GNSS measurements only. An extended Kalman filter with a unified state formulation is developed to process pseudorange and multi-antenna carrier-phase observations, enabling simultaneous orbit and attitude estimation without inertial aiding. The estimated attitude is directly employed as feedback in a quaternion-based nadir-pointing controller, creating a closed-loop interaction between spacecraft dynamics and GNSS measurement geometry. Simulation results demonstrate that while increasing elevation mask angles elevates Geometric Dilution of Precision (GDOP) and monotonically degrades position accuracy, attitude estimation exhibits a constrained degradation. Furthermore, variations in the initial tumbling rate alter the spatio-temporal evolution of measurement geometry, revealing a strongly non-monotonic convergence behavior. These findings highlight the feasibility of minimal-sensor architectures and emphasize the critical coupling between rotational dynamics and estimator observability.