Implementation of Roll Angle and Angular Velocity Estimation Algorithm for a High-Speed Projectile Using Accelerometers Output Data

Abstract

In this paper, implementation of roll angle and angular velocity estimation algorithm for a high-speed projectile using the fusion of the accelerometers output data is proposed. The reason for the use of accelerometers instead of gyros and magnetometer is the high error of the MEMS gyroscope for high speed and the low accuracy of the magnetometer due to the presence of Non-Earth magnetic fields and the effects of hard and soft iron. After expression of the proposed algorithm, the implementation process is explained. In this process, an electric motor is used to simulate the projectile roll and two accelerometers are used to measure angular velocity and acceleration. Two constant and variable velocity scenarios have been investigated in both online and offline modes. Both extended Kalman Filter and adaptive extended Kalman filter estimators have been used to estimate the rolling angular velocity. Finally, the comparison of these two methods for the rolling angular velocity and roll angle, indicates a better performance for the adaptive estimator.