21 September 2022 in IEEE Sensors Journal
According to the problems that the velocity of OD(Odometer, OD) and DVL (Doppler Velocity Log, DVL) is easily contaminated by non-Gaussian noise, and the priori information of state vector is usually hard to estimate accuracy enough, which will lead performance of alignment to decline or even divergent. Considering the error characteristic model has very significant nonlinear characteristics. A RUIKF (Robust Unscented Information Kalman Filter, RUIKF) algorithm for dynamic initial alignment is proposed in this paper to suppress the influence of interference noise. The RUIKF algorithm firstly calculate the MD(Mahalanobis distance, MD) between observation and the innovation of observation, and an amplification factor is designed to modify the estimation of the measurement noise covariance matrix, and then take the modified the measurement noise covariance matrix to update measurement. The state error is estimated by RUIKF to modify the SINS (Strapdown inertial navigation system, SINS) navigation solution. Initial alignment experiments are carried out by RUIKF and UIKF (unscented information Kalman filter,UIKF) algorithm respectively under the conditions of outliers and mixed Gaussian distribution noise. The results show that RUIKF can suppress the effect of outliers and non-Gaussian noise when the velocity of OD(or DVL) is contaminated. The experiment results also verify the feasibility and effectiveness of the RUIKF algorithm on SINS/DVL/OD integrated dynamic initial alignment.
17 November 2025 in IEEE Transactions on Instrumentation and Measurement
High precision marine gravity field construction is the basis of gravity aided inertial navigation, which requires gravity measurement with high precision and resolution. Shipborne atom gravimeter dynamic measurements satisfythese characteristics, but the output sequenceisseriously disturbed by vibration noise, so noise reductionshould be carried out. We propose a noise reductionmethod, in which the sequenceis decomposed into different frequencydominatedmodes by successive variational modedecomposition (SVMD), and the mode is judged as signal or noisedominatedby spacedependent recursive sample entropy(SdrSampEn). Then the Savitzky-Golay(SG)filter is used to smooth thesignaldominatedmode, and singular spectrum analysis (SSA) is used to extract weak useful signal from the noise dominatedmode, and the denoised sequenceis reconstructed. The noise reduction performance of the proposedmethod was verified bytwo groups of dynamic measured data of atom gravimeter, which improved at least 38.3% compared to existing methods. The method was applied to the accuracy evaluation of atom gravimeter survey line, and the sub-mGal internal and external coincidence accuracy are obtained.