A new method of GNSS fault data detection for strapdown land vehicle gravimetry

Affected by complex observation environment in land vehicle gravimetry, the use of Global Navigation Satellite System (GNSS) which is playing a key role has been seriously challenged currently. A new GNSS Fault Data Detection approach for land vehicle gravimetry is proposed in this paper. Taking advantage of the property that Strapdown Inertial Navigation System (SINS) can maintain the high precision positioning result in a short period of time, the abnormal GNSS data can be detected by Kalman filtering method which uses the difference between the position and velocity of the two systems (GNSS and SINS) as the observation information. After fixing the GNSS fault data, the result of land vehicle gravimetry will also be improved accordingly. Applying this method in a typical vehicle gravity measurement test with SGA-WZ02 strapdown gravimeter, the accuracy of internal coincidence of the four repeated measure lines improved from 0.65 mGal to 0.55 mGal, while the external accuracy of four measure lines improved from 1.29 mGal to 1.24 mGal. Practical gravimetry result indicates that the method proposed in this paper can not only improve the GNSS observation data, but also can improve the accuracy of gravity measurement effectively. Finally, some discussions and suggestions are put forward for the applicability of this method and its further improvement.