An Improved Robust Estimation Method for GNSS/SINS under GNSS-Challenged Environment

Abstract

The outlier measurement affects the accurate estimation effect of tightly coupled global navigation satellite system (GNSS) and strapdown inertial navigation system (SINS) integrated navigation parameters under GNSS-challenged environment. To improve the convergence speed and robustness of nonlinear filter used in tightly coupled GNSS/INS under GNSS-challenged environment, an iterated cubature Kalman filter (CKF) based on the improved robust estimation method is proposed. Firstly, the mind of nonlinear least squares regression is included into CKF framework, and multiple iterations are used to improve the convergence speed of the filter and the error compensation effect. Then, a simplified iterated update structure is developed to reduce the computational cost for integrated navigation system. Moreover, the Geman McClure (GM) loss function is introduced to reduce the weight of outlier measurement, which improves the robust estimation ability of the filter. The field experiment indicates that the proposed method has better compensation effect than traditional methods on navigation errors in the case of frequent signal outages.