A Random Error Compensation Method for MEMS Gyroscope Based on Improved EMD and ARMA
It is difficult to increase the measurement accuracy of micro-electro-mechanical system ( MEMS) gyroscope due to random error. An error compensation method based on improved empirical modal decomposition ( EMD ) and an optimized autoregressive moving average ( ARMA ) model is proposed to lessen the random error of MEMS gyroscope. The proposed method is used to extract the noise and trend components from a signal based on Hausdorff distance,the mean of the accumulated standardized modes,and the conventional empirical mode decomposition. Then,ARMA modeling and filtering are applied to the remaining components. The ordering procedure of ARMA model is optimized using the sand cat swarm optimization algorithm. The improved adaptive filter is used to compensate the random error. The experimental results show that,compared to the traditional EMD and traditional ARMA model,the root mean square error obtained by the proposed method in static experiment is decreased by 52.5％ and 34.4％ and the root mean square error obtained by the proposed method in dynamic experiments is decreased by 50％ and 32.35％,respectively. The proposed method might successfully reduce the random error and raise the measurement accuracy of MEMS gyroscope.
micro-electro-mechanical systemgyroscopeimproved empirical mode decompositiontime series modellingHausdorff distanceadaptive filter
万方数据
维普
