未知但有界噪声条件下的MEMS陀螺信号处理方法.

A novel model identification and error compensation technique for micro-electronic-mechanical system (MEMS) gyroscope based on ellipsoidal bounding algorithm is proposed. Firstly, an autoregressive model for stochastic drift is established. For the time-varying characteristics of the MEMS gyroscope error model, a real-time recursive method called bounding ellipsoidal adaptive constrained least-squares (BEACON) algorithm is adopted to realize the dynamic parameter identification and improve the modeling accuracy. Then, a bounding ellipsoidal adaptive state estimation (BEASE) algorithm with unknown but bounded (UBB) disturbances is proposed to estimate the regular rate. The new weighting strategy and optimization criterion are used at observation updating stage. The updating process of the feasible state set and the selection method of the optimal parameters are reduced under this framework. The method is applied to process the MEMS gyroscope signals and the experiment results to verify the efficiency and improved performance of the proposed method. [ABSTRACT FROM AUTHOR]