Improvement of Euler Angle Algorithm Based on Extended Kalman Filtering

Euler angles are essential parameters to describe the attitude of a carrier, and the Euler angles computed by the Inertial Measurement Unit (IMU) are affected by the performance of the components. Traditional Kalman filtering algorithms struggle to effectively suppress nonlinear Gaussian noise in engineering applications. Therefore, this paper employs an improved Extended Kalman Filtering algorithm(EKF), which linearizes the nonlinear system using the Jacobian matrix. It utilizes quaternions and gyroscope errors as state variables, while accelerometers serve as measurement inputs to update the quaternions, thereby obtaining the Euler angles. Building on the updated Euler angles, a magnetometer is incorporated to further refine the calculation, mitigating the impact of gyroscope z-axis drift on the heading angle and eliminating external magnetic interference. Experimental results demonstrate that, compared to the Madgwick algorithm and complementary filtering, the improved algorithm significantly reduces the influence of component errors on Euler angle calculations, minimizes bias drift, and better reflects the actual operational conditions of the carrier, effectively achieving filtering results.