Strapdown Gravity Disturbance Vector Determination Method Based on Error-Coupled Constraints and Direct Differencing-Optimal Estimation Synergy

This article proposes a strapdown gravity disturbance vector (GDV) measurement method based on error-coupled constraints and direct differencing-optimal estimation synergy. Through theoretical derivation and simulations, the coupling mechanism among horizontal attitude errors, accelerometer biases, and GDV is revealed. Building on this analysis, a mathematical model for GDV determination under coupling constraints is established. Incorporating stochastic process modeling of the GDV, a Kalman filtering–smoothing model is formulated in which EGM2008 low-frequency reference values and the preliminary GDV results obtained via direct differencing are assimilated, thereby realizing a synergistic fusion of the direct differencing and optimal estimation methods. These innovations are integrated into a stepwise error separation strategy, yielding a comprehensive strapdown GDV measurement scheme. Airborne gravimetry experiments demonstrate internal consistency accuracies of 1.30, 0.80, and $0.44{\,}\text {mGal}$ for the eastward, northward, and upward components, respectively, with a spatial resolution of $4.8{\,}\text {km}$ . The method significantly enhances separation capability between systematic errors and the GDV, yielding marked improvements in horizontal components measurement precision.