Experimental analysis of quaternion-based attitude estimation algorithms

This paper investigates the performance of quaternion fusion filtering (QFF) and quaternion extended Kalman filtering (QEKF) algorithms on small embedded systems. Firstly, based on the Euler rotation matrix and quaternion rotation matrix, the computational model for the QFF algorithm is provided; Secondly, based on the quaternion state equation and observation equation, the principle of the QEKF algorithm is introduced; Finally, the experiments are designed to analyze the performance of the two algorithms in terms of computational complexity and estimation accuracy. The results show that both attitude estimation algorithms can obtain relatively accurate attitude angles, and the QFF algorithm has less computational complexity and simpler algorithm implementation, while the QEKF algorithm has higher accuracy.