A Mismatch Correction Method Based on Two-Way Strategies of Spatial Vector Feature

Gravity matching method is the core technology in gravity aided inertial navigation system (GAINS). The single point filter matching algorithm is widely used in underwater navigation owing to its high accuracy and real-time performance. However, due to gravity measurement errors, the mismatch points may occur and cause large position errors. In order to improve the matching accuracy, a mismatch correction method based on two-way strategies of spatial vector feature is proposed. Based on the spatial vector characteristics of the gravity field, two-way strategies of scale constraint and direction constraint is applied to detect the outliers. Vector median filtering correction method is applied to calculate the accompanied matching points. The affine transformation method with least squares optimization parameters is adopted to smooth the corrected trajectory. Simulation results show that the proposed method can effectively suppress the spatial position error and improve the position accuracy.