As the Global Navigation Satellite Systems (GNSS) are intensively used as main source of Position, Navigation and Timing (PNT) information for maritime and inland water navigation, it becomes increasingly important to ensure theudreliability of GNSS-based navigation solutions for challengingudenvironments. Although an intensive work has been done inuddeveloping GNSS Receiver Autonomous Integrity Monitoringud(RAIM) algorithms, a reliable procedure to mitigate multipleudsimultaneous outliers is still lacking. The presented workudevaluates the performance of several methods for multipleudoutlier mitigation based on robust estimation frameworkudand compares them to the performance of state-of-the-artudRAIM methods. The relevant methods include M-estimation,udS-estimation, LMS and RANSAC-based approaches as well asudcorresponding modifications for C/N0-based weighting schemes.udThe snapshot positioning methods are also tested within theudquaternion-based Cubature Quadrature Kalman filter forudintegrated inertial/GNSS solution. The presented schemes areudevaluated using real measurement data from challenging inlandudwater scenarios with multiple bridges and a waterway lock. Theudinitial results are encouraging and clearly indicate the potentialudof the discussed methods both for classical snapshot solutions as well for the methods with complementary sensors.
展开▼
