Roll Angle Estimation for Smart Munitions

Abstract

The roll angle is estimated only using the pitch and yaw rates for a gunlaunched smart munition with fast rolling of 1~2 Hz. An extended Kalman filter is designed based on the roll, pitch, and yaw dynamics and two MEMS gyroscopes measuring pitch and yaw rates. It is simulated under a realistic condition with low-grade gyroscopes. The simulation results show that the roll angle is estimated with an accuracy of 1~3 deg. (1-sigma error). The filter also exhibits that it is robust to the scale factor and bias errors of the MEMS gyroscopes. And the proposed algorithm is experimented using a two-axis rate table and a low grade MEMS IMU. The results show that the algorithm can be applied to an actual system.