Improved Complementary Filters for Estimation of Heading and Attitude Angles in Accelerated Motion

In this paper, a method for improving the performance of Complementary filter in the Attitude and Heading Reference System for estimating the orientation in accelerated movements is presented. Although existing complementary filters have advantages such as low computational volume, stability in different dynamic conditions, effectiveness at low sampling rates, and simplicity in the parameter setting process; But in the situation where the mobile device is exposed to non-gravitational accelerations, they show inappropriate performance. The proposed algorithm is designed based on the threshold-based path selection method and by adjusting the gain of complementary filters according to the size of the external acceleration, it improves the estimation of angles. In the following, the proposed algorithm is compared with the Extended Kalman filter and its three adaptive versions. The simulation and evaluation results of the proposed method show that the improved complementary filters achieve good performance in accelerated movements compared to the AEKF in the direction and alignment reference system.