Research On Error Compensation Of Random Drift Based On MEMS Gyroscope

MEMS(Micro Electromechanical System)gyroscope.has the advantages.of small size,long.life,low cost,high.impact and low power.+consumption.characteristics,wid-ely used in many fields,such as car navigation,robot pose measurement system,cam-era image stabilization platform,virtual somatosensory games,electronic toys,etc.In addition,after no foreseeable such as UAV detection equipment and other weapons systems,inevitably.toward.miniaturization,intelligent,digital and high mobility direction,so the MEMS gyroscope has great development value and broad prospects,its excelle-nt characteristics make it widely concern all over the world,and has been listed as one of the key technologies of the revitalization and development of the twenty-first Century.However,the.accuracy of MEMS.gyroscope is.relatively low,making it the bottleneck of the development of guidance and control system,micro navigation system and other fields.There are two ways to improve the accuracy of MEMS gyroscope.First,improve the performance of the system from the hardware structure,secondly,the point of view of the.algorithm can effectively..reduce.the MEMS gyroscope.random.drift error,improveing the.measurement accuracy.From the point of view of software,this paper.shows the noise.characteristics of.MEMS gyroscope and the random drift error compensation technology of MEMS gyroscope.In this paper,several performance indexes and internal working principle of the gyroscope are introduced firstly,and the Allan variance method is used to analyze the noise characteristics of the MEMS gyroscope.After the system introduces the theory of time series modeling method,in order to enhance the reliability of the test data,the special collection of 10 groups of data by using Rajda criterion removing singular point method for gyro output data preprocessing,then with stepwise regression method was used to fit the trend of drift,convert it to zero mean stationary data.Finally,the AR model is established by calculating ACF and PCF.Then the Kalman filter is used to deal with the random drift error of 10 groups of MEMS gyroscope.The experimental results show that the Kalman filter based on time series is effective in the error compensation of MEMS gyroscope random drift.The dynamic error processing of MEMS gyroscope,on the turntable as the calibration set,the MEMS gyroscope for the different rates of uniform motion,the classic.Kalman filter to.solve the.dynamic.err-or,is still feasible and effective.When the MEMS gyroscope is in motion,an adapti-ve Kalman filter is designed to suppress the dynamic drift of the MEMS gyroscope and improve the accuracy of the MEMS gyroscope.