A combined CKF-PSR method for random noise compensation of vibratory gyroscopes

Abstract

The micro-electromechanical system (MEMS) vibratory gyroscope has been widely used in many fields, such as aerospace, autonomous vehicles, and robotics. The major affecting factor of the final accuracy of this type of gyroscope is random noise error. The MEMS vibratory gyroscope has been a hot research topic in the vibratory gyroscope field. This paper considers the vibratory gyroscopes, such as metal resonant and tuning forks. In the process of research and analysis, the MEMS gyroscopes are mainly used. Aiming at suppressing the random noise of the MEMS gyroscopes effectively, a nonlinear suppression method based on the Cubature Kalman Filter-Phase Space Reconstruction (CKF-PSR) is proposed. After fully analyzing the data flow and processing timing of the MEMS gyroscope, the chaotic attractor PSR is used to complete the data modeling of the gyroscope output. Combining the established model with the nonlinear factors of the gyroscope output, a nonlinear filter is proposed for comprehensive filtering. The algorithm complexity and running dynamics are systematically analyzed and verified. Finally, it was applied to a self-developed inertial integrated navigation device. The results show that compared to the traditional filtering methods, the Finite Impulse Response (FIR) and the Kalman Filter - Phase Space Reconstruction (KF-PSR), the proposed method has significant advantages, laying a theoretical foundation for the application of vibratory gyroscopes in high-precision applications.