A Computationally Efficient Outlier-Robust Cubature Kalman Filter for Underwater Gravity Matching Navigation

Abstract

Gravity-aided navigation is one of key techniques for the navigation of underwater vehicles. Cubature Kalman filter (CKF)-based matching algorithm improves the positioning accuracy of traditional Sandia inertial terrain aided navigation method. However, the gravity sensor may suffer from outlier interferences due to complex and changeable underwater environments, which degrades the performance of CKF-based matching algorithm remarkably. To address this problem, in this article, a novel computationally efficient outlier-robust CKF-based matching algorithm is proposed for an underwater gravity-aided navigation system with outlier-contaminated measurements. The convergence analysis and the stability discussions are given to show the effectiveness of the proposed algorithm, and the discussion on computational complexity illustrates the good real-time performance of the proposed algorithm. Simulation and experimental results demonstrate the advantages of the proposed matching algorithm as compared with existing state-of-the-art matching algorithms.