GNSS Satellite Fault Detection and Exclusion for Integrated GNSS/INS Systems

Abstract

In this work, we present a Global Navigation Satellite Systems (GNSS) satellite fault detection, identification, and exclusion method for an integrated GNSS/Inertial Navigation System (INS) tightly-coupled navigation architecture. Observed satellite fault profiles and magnitudes are taken as a baseline to design the monitor. Normalized innovations for each satellite are utilized to observe the respective satellite faults on each channel. A sufficient test statistic for each satellite channel is derived using the generalized likelihood ratio test (GLRT). We develop a method for quantifying missed detection rates given fault profile and magnitude. This developed methodology is capable of providing protection levels using the INS-only coasting solution. We also describe an identification and exclusion method using consecutive windows of the monitor without the need for running parallel Kalman filters (KF). We validate the developed algorithm for the detection, identification and exclusion of faulty satellites using simulated fault scenarios.