A signal processing technique for compensating random drift of MEMS gyros

In this paper, we present a prediction and compensation method for Micro-Electro-Mechanical System (MEMS) gyroscope random drift, which is based on relevance vector machine. The relevance vector machine (RVM) model is established based on the feature of MEMS gyroscope random drift and the parameters are trained by the Expectation Maximization (EM) algorithm. By phase space reconstruction, the time sequence of random drift is accessed in the model. The final experimental results indicate that our proposed methodology can achieve both the least complexity of structure and goodness of fit to data, and also can predict the gyroscope random drift accurately. Furthermore, by compensating random drift using the predicting result, the precision of gyroscopes application could be improved well.