Separability Analysis for Multiple Faults in GNSS/INS Integration

In the integration of GNSS and inertial sensors, any potential sensor failures or faulty measurements due to harsh operating environments or malfunctions of sensor components may result in poorly estimated navigation parameters. If the sensor component failures or faulty measurements occur and cannot be detected immediately, overall system performances will de degraded. Thus, real time detection of failures in estimations system is critically important, particularly for some safety or liability-critical applications. Therefore, it is certainly necessary to implement an online fault detection and isolation (FDI) algorithm. Various approaches of fault detection and isolation have been proposed and implemented, and at the same time, the Minimum Detectable Biases (MDBs) in the GNSS/INS integration and the impact of these MDBs on the estimated navigation parameters can also be defined. However, in order to identify and remove the specific faulty measurements, the system should have the ability to separate these faults. This paper presents the latest separability analysis for multiple fault scenarios within the GNSS/INS integration, which is of importance in developing future robust positioning and navigation systems with the fault detection and isolation procedures.