Attitude Estimation Based on Angular Accelerometer in High Dynamic Environments

This paper addresses the challenge of attitude estimation drift in high dynamic environments by proposing an attitude algorithm based on angular accelerometer. The system configuration includes an angular accelerometer, a linear accelerometer, and a magnetometer. A mathematical model using the rotation matrix as the state variable is developed, and the invariant unscented Kalman filter is employed for attitude estimation. Through simulation experiments, the accuracy of the proposed algorithm is compared with a traditional gyro-only method. Results demonstrate that the proposed algorithm offers superior performance in high dynamic scenarios, significantly enhancing system bandwidth and reducing error drift. This research contributes to improving precision in high dynamic attitude estimation applications.