Kinetic Model-Constrained Robust In-Motion Alignment for Projectiles SINS/GNSS

Rapid and accurate initial alignment is a critical prerequisite for the performance in integrated navigation of strapdown inertial navigation system (SINS) and global navigation satellite system (GNSS), particularly in guided projectiles operating under in-motion conditions. However, conventional optimization-based in-motion alignment (OBIA) methods often suffer from slow convergence and reduced accuracy in the presence of degraded GNSS signals and the cumulative errors of low-cost micro-electro-mechanical system (MEMS)-based inertial measurement unit (IMU). To address these challenges, this paper introduces a novel kinetic model-constrained robust in-motion alignment approach tailored for projectile applications. The proposed method integrates a simplified ballistic trajectory model into an extended Kalman filter (EKF) to constrain GNSS measurements, while innovation-based anomaly detection and an adaptive filtering strategy enhance robustness against GNSS outliers. Furthermore, a sliding-window integration scheme is employed to suppress drift errors from the MEMS-based IMU. Semi-physical simulation experiments under both nominal and degraded GNSS conditions demonstrate that the proposed method significantly outperforms conventional OBIA techniques in terms of convergence speed, alignment accuracy, and resistance to interference. These findings validate the proposed approach as an effective solution for rapid and robust initial alignment in time-critical projectile navigation scenarios.