Akademik Veri Yönetim Sistemi
22nd International Technical Meeting of the Satellite Division of the Institute-of-Navigation (ION GNSS-09), Savannakhét, Laos, 22 - 25 Eylül 2009, ss.1965-1973, (Tam Metin Bildiri)
The performance of the estimation algorithms used in aided navigation applications is significantly affected by the accuracy of the error models used for inertial sensors. Despite its profound importance, a standard procedure for modeling inertial measurement unit errors has yet to be developed. In this study, a new AR modeling method based on wavelet decomposition is presented for MEMS based IMUs. The wavelet decomposition is used to compute the scaling coefficients from which the AR model parameters are extracted. In order to remove the effects of changing ambient temperature on error modeling process, a new method to compensate for the temperature effects is also introduced. The laboratory test results of the proposed method verified that our procedure is quite successful in deriving inertial sensor's stochastic error models which can be directly used in Kalman filtering applications. Accurate inertial sensor error models obtained with this method will lead to improvement of the navigation filters' performance especially for MEMS based systems