Vibration-Robust Attitude and Heading Reference System Using Windowed Measurement Error Covariance

In this paper, a new technique for attitude and heading reference system (AHRS) using low-cost MEMS sensors of the gyroscope, accelerometer, and magnetometer is addressed particularly in vibration environments. The motion of MEMS sensors interact with the scale factor and cross-coupling errors to produce random errors by the harsh environment. A new adaptive attitude estimation algorithm based on the Kalman filter is developed to overcome these undesirable side effects by analyzing windowed measurement error covariance. The key idea is that performance degradation of accelerometers, for example, due to linear vibrations can be reduced by the proposed measurement error covariance analysis. The computed error covariance is utilized to the measurement covariance of Kalman filters adaptively. Finally, the proposed approach is verified by using numerical simulations and experiments in an acceleration phase and/or vibrating environments.