Error modeling and analysis for random drift of MEMS gyroscopes

In order to comprehensively understand the characteristics of random drift error of micro-electro-mechanical systems (MEMS) gyroscope varying with time, the dynamic Allan variance was utilized to analyze the property of output signal of MEMS gyroscopes. Firstly, the theory of Allan variance and dynamic Allan variance were introduced. Then, the Allan variance and the dynamic Allan variance were utilized to analyze the characteristic and performance of the practical testing date of MEMS gyroscopes respectively. The study results show that the main random noise of MEMS gyroscope are the rate ramp, quantization noise and the rate random walk, and they are unstability varying with time. The dynamic Allan variance was utilized to not only extract and confirm some main noises of MEMS gyroscope, but also track and describe the stability of MEME gyroscope's signal varying with time. Consequently, compared with Allan variance, the dynamic Allan variance can represent the characterization of stochastic errors in MEMS gyroscope more comprehensively.