The performance of MEMS gyroscope is crucial to its application, and the zero-bias stability is the main characteristic parameter of gyroscope performance. In this paper, a method of using Allan variance is presented. Experimental results using this method with gyroscopes of different precision show that Allan variance is a reasonable tool to analyze the performance of MEMS gyroscopes. Through analyzing Allen variance, the performance of MEMS can be estimated with precision and a large amount of data is not necessary as with standard variance.

MEMS Gyroscope Performance Estimate Based on Allan Variance

Proper selection of the dither amplitude and frequency of the ring laser gyroscope (RLG) influences the effect of dither bias directly. In order to reduce RLG lock-in effectively, the dynamic lock-in width is obtained by deducing the basic RLG input-output equation in the article. The corresponding optimization model is based on the purpose of reducing the influence of dynamic lock-in. The result of the optimization shows that it is necessary to enlarge dither depth to reduce the nonlinear influence. When dither depth is large enough, there is no evident effect on the reduction of dynamic lock-in by enlarging it further. However, dynamic lock-in may be reduced to zero by modulating dither depth locally with very small value according to the results of optimization. Some constructive conclusions are obtained at the end of the article.

Optimization of dither amplitude and frequency of RLG

A base motion isolation algorithm is proposed for a rate biased RLG (Ring Laser Gyro) north finder subjected to vehicle disturbances. A single-axis rotation modulation scheme is used. Furthermore, the angular increment of the rotation of the turn table is utilized to compute the outputs of the third virtual gyroscope, by which the north finder consist of two gyroscopes and three accelerometers are translated into a Strapdown Inertial Navigation System (SINS) consist of three gyroscopes (two real and one virtual) and three accelerometers. Therefore, the north-finding problem is translated into an initial alignment problem based on a Kalman filter. The experiments results show that the new algorithm can effectively avoid the influences of vehicle disturbances. The 20min north finding accuracy of azimuth angle is better than 1′ when the weather is gusty, the vehicle is rocking and the motor is running on.

Base motion isolation algorithm for rate biased RLG north finder with disturbances

The initial alignment of laser gyroscope inertial navigation system (LINS) for land vehicles is difficult when disturbance are present, for example when the motor is running on or people are walking in the vehicle The Kalman filter alignment technique was developed for signal analyses of the LINS inertial instruments The observation signals were pre processed to reduce the interference, then the Kalman filter algorithm was improved to increase the stability The experimental results show that this technique effectively increases the alignment accuracy with disturbances

Initial alignment of laser gyroscope inertial navigation system with disturbances

The control electronics and the signal processing for ring laser gyro (RLG) influence its accuracy and other performances, such as the dynamic capability etc In order to satisfy the need for the inertial measurement unit (IMU), the methods to improve the accuracy and dynamic capability of L 1 RLG are researched A new developed control electronics for the L 1 RLG is introduced, a high frequency sampling system for the RLG is realized on the basis of this electronics, and a method of digital signal processing for filtering dither motion is investigated As shown by the experimental results, the accuracy of the L 1 RLG is improved, and the sampling frequency is increased

Research on dither control and signal readout of ring laser gyros

Errors of inertial navigation systems are mainly caused by the errors of the gyros. The evaluation and compensation of errors of gyros is an important way to improve the precision of gyros. It is necessary to build the random error model of gyros. Several methods can be used to analyse random error model of ring laser gyro (RLG), such as power spectrum density (PSD), ARMA, and Allan variance. Allan variance is a tool to analyse the stability of frequency in the time domain. There is quantitative relationship between Allan variance and power spectrum density of noise. Various kinds of ring laser gyro noise can be extracted from the gyro data with Allan variance. It is simpler than to analyse PSD in frequency domain. L 1 RLG was analysed with Allan variance, and the errors which exist in L 1 RLG were given. It is shown that Allan variance is a practical tool to build the random error model of ring laser gyros.

Teng Yunhe

Papers

Initial alignment of laser gyroscope inertial navigation system with disturbances

Research on dither control and signal readout of ring laser gyros

Investigation on random error model of ring laser gyro