An effective unscented Kalman filter for state estimation of a gyro-free inertial measurement unit

Abstract

This study reports a gyro-free inertial measurement unit (IMU) using solely four triaxial accelerometers. System equations and a configuration which is feasible for the gyro-free IMU design are presented. The propagation of accelerometer measurement errors is analyzed. An unscented Kalman filter (UKF) is proposed for state estimation. Simulation results show that the system state is robustly estimated by the proposed UKF. Furthermore, compared with the results of error analysis, the UKF provides effective error reductions on state estimation. The error of angular velocity estimation over full scale (FS) is about ±0.4%FS.