During high dynamic gravity measurements conducted on unmanned surface vehicle, the presence of low-frequency noise caused by the vertical and horizontal motion disturbances of the carrier in conjunction with the low-frequency excitation noise from the sensor, results in a direct mixture within the frequency band of the gravity signal. Admittedly, conventional filtering techniques such as finite impulse response (FIR) or infinite impulse response (IIR) filtering prove insufficient in eliminating the measurement noise, ultimately leading to a decrease in gravity measurement accuracy.In this regard, this paper proposes the use of the kalman smoothing method as a replacement for the traditional frequency domain low-pass filtering technique. This method allows for the identification of gravity anomaly information even in the presence of noise by employing optimal estimation methods.Given that the gravity measurement data is processed offline, this paper further utilizes the optimal fixed interval smoothing algorithm to process the gravity measurement data obtained from unmanned surface vehicle. This algorithm enhances the accuracy beyond what is achievable with traditional frequency domain low-pass filtering techniques. To validate the effectiveness of our proposed algorithm, we have conducted processing on real sea test data, confirming its efficacy.
|