Study on Extended Kalman Filtering for Attitude Estimation of Micro Flight Vehicle

In this paper an extended Kalman filter for real-time estimation of micro flight vehicle body attitude by using MEMS gyros, accelerometers, and magnetometers is presented. The algorithm uses quaternions to represent rotation, which can avoid the singularity problem caused by using Euler angles. Attitude quaternion errors and drift bias of gyros were selected to construct a state vector. The state equation is established on the base of attitude quaternion error differential equation. The measurements of Kalman filter were obtained through iteration using acceleration and magnetic field vector. The algorithm can significantly reduce calculated amount and can converge quickly to rotation attitude.