Improved filter estimation method applied in zero velocity update for SINS

Zero Velocity Update (ZUPT) utilizes the zero velocity condition for stationary Strapdown Inertial Navigation System (SINS), executes navigation errors estimation and emendation to control SINS position growth. The improved filter estimation ZUPT is proposed in this paper. Two technologies, separate-bias Kalman filter and yaw error rapid estimation, are applied in the proposed method. Separate-bias Kalman filter solve the divergence problem for setting a mass of Kalman filter state variances. And the yaw error rapid estimation can calculate the unobservable yaw error in periodic ZUPT. Simulation results show that the estimations of attitude errors, yaw error and position errors converge to the right values quickly in the proposed ZUPT, and the position accuracy is improved effectively compared to the conventional one.