Work

A Two-Stage Airborne Vector Gravimetry Method Based on Error Decoupling and Frequency-Division Correction

Year: 2026

Type: article

Abstract: High-precision gravity vector data are essential for geological structure interpretation and resource exploration. Strapdown airborne vector gravimetry is the mainstream technique for efficient medium-to-high frequency gravity field acquisition. However, closed-loop coupling between horizontal gravity disturbance and attitude error has long constrained its horizontal accuracy. To tackle this limitation, we deduce the multi-source error propagation law of strapdown gravity vector measurement, and quantitative analysis confirms closed-loop coupling as the core factor degrading horizontal precision. On this basis, we propose a two-stage airborne vector gravimetry method based on error decoupling and frequency-division correction. In the error decoupling stage, we design a gyro-standalone attitude update (GSAU) framework to isolate the reverse error transmission channel where gravity disturbances contaminate attitude computation, achieving mechanistic decoupling between attitude errors and gravity signals. In the error correction stage, a tapered-band division complementary filtering (TBDCF) module is constructed for frequency-division correction, which suppresses Gibbs ringing artifacts induced by traditional frequency-domain correction and realizes measured gravity data calibration constrained by EGM2008. Airborne gravimetry experiments show that at 3.6 km half-wavelength resolution, the internal consistency accuracies reach 0.77, 0.92 and 0.45 mGal for east, north and upward components respectively. Relative to the conventional scheme, horizontal and upward internal consistency accuracy gains exceed 66% and 33%, respectively. Requiring only mechanism-level optimization and algorithmic improvements, the method needs no extra hardware and integrates directly into existing strapdown airborne gravimetry systems.

Source: IEEE Transactions on Geoscience and Remote Sensing

DOI: 10.1109/tgrs.2026.3751811