The purpose of this study is to explore a signal processing method to improve the angular rate accuracy of micro-electro-mechanical system (MEMS) gyroscope by combining numerous gyroscopes.
To improve the dynamic performance of the signal processing method, the interacting multiple model (IMM) can be applied to the fusion of gyroscope array. However, the standard IMM has constant Markov parameter, which may reduce the model switching speed. To overcome this problem, an adaptive IMM filter is developed based on the kurtosis of the gyroscope output, in which the transition probabilities are adjusted online by utilizing the dynamic information of the rate signal.
The experimental results indicate that the precision of the gyroscope array composed of six gyroscopes increases significantly and the kurtosis-based adaptive Markov parameter IMM filter (K-IMM) performs better than the baseline methods, especially under dynamic conditions. These experiments prove the validity of the proposed fusion method.
The proposed method can improve the accuracy of MEMS gyroscopes without breakthrough on hardware, which is necessary to extend their utility while not restricting the overwhelming advantages.
A K-IMM algorithm is proposed in this paper, which is used to improve the angular rate accuracy of MEMS gyroscope by combining numerous gyroscopes.
Shen, Q., Liu, J., Huang, H., Wang, Q. and Qin, W. (2017), "Kurtosis-based IMM filter for multiple MEMS gyroscopes fusion", Sensor Review, Vol. 37 No. 3, pp. 237-246. https://doi.org/10.1108/SR-08-2016-0147Download as .RIS
Emerald Publishing Limited
Copyright © 2017, Emerald Publishing Limited