Research of unscented Kalman filter application in polarization/MEMS integrated navigation system

MEMS-IMU usually needs to be integrated with other navigation methods because its output error accumulates over time. Integrating polarization navigation sensor with MEMS-IMU in a system is an effective method to improve the navigation performance. Because the yaw angle of polarization sensor is not divergent, the data can be used to adjust output of MEMS-IMU. Car experiments have been done to examine the viewpoint. In addition, this paper proposes that UKF can also enhance the performance of the system. The result shows that drift generated by MEMS-IMU can be restrained with assist of polarization sensor. The navigation information reckoned is close to the real through the compensation of UKF.