A Comparison of Two Approaches to SINS Initial Alignment for Marching Vehicle

Classic Kalman filtering algorithm needs a little precise initial value, but coarse alignment precision of SINS for marching vehicle can't ensure meet the requirement in all working conditions. UKF(unscented kalman filter) was proposed to achieve initial alignment of SINS for marching vehicle with DTU(distance transfer unit) aiding. The system equation of error model based on large misalignment angles was expounded, observation equation was derived when the measurement variables were chosen as difference of velocity offered by SINS and velocity reckoned by DTU, and error model was simplified under the situation of small misalignment angles. UKF algorithm based on additive noise model was expounded detailed. Simulation based on vehicular tests data showed that UKF algorithm could achieve initial alignment for marching vehicle, and is more robust to initial value than classic kalman filtering algorithm.