A Modified Nonlinear Two-Filter Smoothing for High-Precision Airborne Integrated GPS and Inertial Navigation

Abstract

Airborne remote sensing imaging depends on the integrated system of strapdown inertial navigation system (SINS) and Global Positioning System (GPS) to obtain high-accuracy motion parameters. In this paper, a modified nonlinear two-filter smoother (TFS) is proposed for an offline SINS/GPS integrated system suitable for remote sensing imaging. The proposed smoother has a two-filter structure, which includes a forward filter based on central difference Kalman filter, a backward filter with modified propagation and update equation, and a smoothing algorithm. The smoothing algorithm with the modified backward filter is conducted by the simulation and the data processing of SINS/GPS integrated system flight test. Furthermore, a digital camera is used to verify the precision of practical applications in a check field with numerous reference points. In these tests, the performance of the proposed smoother is compared with the central difference Rauch-Tung-Striebel smoother (RTSS), extended TFS, and extended RTSS.