Technique Study On Noise Detecting And Analysis Of Micro-machined Gyroscope

As the core part of the inertial navigation system,the gyroscope has a direct influence on the accuracy and safety of navigation.It has become one of the hotspots of micro-machined inertial sensor.In this paper,the self-developed ball-disc rotor-based gyroscope by Harbin Institute of Technology MEMS Centor is taken as the object.The existed noise source of the gyroscope is analysed,and provides a reference for further improving the gyro precision.First,the paper describes the working principle of each part of the whole gyro system.On this basis,the influence of the error of gyro mechanical structure on the gyro output in machining and assembling process is analyzed in detail.The existence of mechanical structure error is verified,and the method of suppressing the mechanical noise is put forward.Secondly,the random error characteristics of rotor micro-gyroscope are studied based on the overlapping Allan variance.The basic principle of Allan variance and the statistical properties of the main random error source are introduced.The noise figure of the main noise source in the rotor gyroscope is characterized and identified by the least squares fitting variance.The contribution of each noise source to the total noise of the system is reflected by the result,and the iterative method is used to deal with the negative coefficient term in the fitting result.Finally,wavelet threshold denoising method is used to denoise the static data of micro gyroscope.By comparing the experimental results,wavelet analysis algorithm was used.Db-5 was selected as the wavelet basis,and the scale of decomposition is 6.The threshold was selected as soft with “heursure” criteria.The result shows that the method is simple and feasible,which can suppress the random error of the gyroscope effectively.