In recent years,as a very important part of inertial navigation technology,MEMS gyroscope due to its low cost,small size,light weight,high integration and so on,has been widely used in the fields of inertial navigation,industrial control and electronic products.Although compared with the traditional gyroscope,MEMS gyroscope has many advantages,its measurement accuracy is relatively low because of the manufacturing process and design level and other reasons that can not meet the actual needs,which has greatly restricted the development and application of MEMS gyroscope.Especially for the random drift error of gyroscope,because its formation mechanism is very complex,there is no clear rule and it changes with the external environment,so that it can not be compensated and corrected by the conventional method and completely eliminated,which is the main reason that limits the accuracy of MEMS gyroscope.Therefore in this paper,based on the gyroscope in the commonly used low cost MEMS inertial device MPU-6050,the modeling analysis and correction techniques of random drift error of MEMS gyroscope are studied.Firstly,in order to acquire the static drift data of gyroscope in MPU-6050,the MEMS gyro error data acquisition system is designed and constructed with STM32F103C8T6 as the core processor.Secondly,based on the investigation and analysis on the research status and development trends of modeling and filtering technology of random drift error in MEMS gyroscope,the formation mechanism of static drift error and the composition of random drift error of MEMS gyroscope is studied,and the static drift error of the MEMS gyroscope has been identified by Allan variance method.The error analysis results show that the static drift error of the gyroscope is mainly composed of zero bias instability,rate random walk and rate rampThirdly,the static drift data of MEMS gyroscope are pretreated and tested to obtain steady and random error data.Then the AIC criterion is applied for model-order determination and the Yule-Walker equation is used to determine the parameters of model.On this basis,the time series model of random drift error of gyroscope is established,and the Kalman filter is designed combined with the error model to filter the error data.Comparing parameters before and after filtering,the variance of the random drift error data of the MEMS gyroscope after Kalman filtering is decreased to 11.7% of that before filtering.Among the main random errors that affect the accuracy of the MEMS gyroscope,the noise coefficient of zero bias instability,rate random walk and rate ramp is reduced by 68.6%,67.7% and 68.6% respectively,which shows that the Kalman filtering can effectively reduce the random drift error of MEMS gyroscope.Finally,according to the shortage in random drift data processing of MEMS by time series and Kalman filtering,the Singer motion model is used to establish the model of random drift error of MEMS gyroscope,and the particle-Kalman combination filtering method is applied for the data processing of random drift error of gyroscope after zero equalization.Comparing parameters before and after filtering,the variance of the random drift error data of the MEMS gyroscope after particle-Kalman combination filtering is decreased to 1.2% of that before filtering.Among the main random errors that affect the accuracy of the MEMS gyroscope,the noise coefficient of zero bias instability,rate random walk and rate ramp is reduced by 72.8%,76.2% and74.6% respectively,which shows that the particle-Kalman combination filtering can significantly suppress the random drift error of MEMS gyroscope.The experimental results show that time series modeling and Kalman filtering,Singer model based modeling and particle-Kalman combination filtering can both effectively suppress the random drift error of MEMS gyroscope.And the filtering effect of particle-Kalman combination filtering is better than of the Kalman filtering.The research results can effectively correct the random drift error of MEMS gyroscope,and it has certain practical value to improve the navigation accuracy of the inertial navigation system which uses low cost MEMS gyroscope as the main inertial sensor. |