ANALYSIS OF THE ERRORS OF MEMS ACCELEROMETERS BY THE ALLAN VARIATION METHOD

In the article we obtained the generalized equation of errors of SINS accelerometers, its components give a characteristic of the influence of conversion coefficients errors, the inaccuracy in the installation of accelerometer measuring axes, the size effect and the displacement of accelerometer zeroes, and a block diagram of the formation of errors in the output signal of accelerometers is also made. The components of the random error of accelerometers, such as bias instability and acceleration random wolk, as well as due to the presence of white and exponentially correlated (Markov) noise, are analyzed. The models of forming the components of the random error of accelerometers from white noise and white sequence are given. The EMS accelerometers of Gemini CAS211 / 291 type for each axis of the accelerometer with a sampling frequency of 100 Hz are investigated, as a result Allan variance graphs and the values of bias instability and velocity and acceleration random walk for each axis are determined. It was found that in the output signals of the accelerometers there is no Markov noise, and the Z axis is noisier than the X and Y axes.