Research on integrity architecture of GNSS/INS integrated navigation systems

Abstract

This study developed an integrity architecture for a Kalman filter (KF) based Global Navigation Satellite System (GNSS)/Inertial Navigation System (INS) integrated navigation system to assure integrity of unmanned aircraft system (UAS) navigation systems. An integrity risk allocation tree is designed for each sensor fault hypothesis including a nominal hypothesis, a GNSS fault hypothesis, and an INS sensor fault hypothesis. KF vertical protection levels (VPLs) were derived for each sensor fault hypothesis, including the INS fault hypothesis to provide complete integrity assurance for integrated UAS navigation systems.

A KF is a recursive estimator that utilizes past measurements to generate the current state estimates, unlike a least-square (LS) filter that is a snapshot estimator. Hence, position estimates as well as sensor fault impacts recursively propagate through KF algorithms. This recursive process should be represented within fault monitoring algorithms and in the formulation of the VPL equations. The architecture exploits existing GNSS integrity methods to guarantee the required levels of integrity for GNSS. Based on the monitored GNSS measurements, potential INS sensor faults are monitored using the KF innovations under the assumption of no GNSS fault. Unlike the measurement sequence fault case, it is possible to fully capture the fault impact on the user position state in real-time using a KF innovation vector under INS fault conditions. Exploiting this fact, a VPL was mathematically derived utilizing the KF innovation with additional uncertainty noise bounding terms which bound a potential user position with a level of system integrity.

Performance evaluations were conducted by simulating KF VPLs based on different types of GNSS/INS integrated systems. Based on these evaluation results, an operational concept for UASs flying within the designed airspace is suggested. As the VPLs that integrated navigation systems can achieve get smaller, the allowed flight region could be enlarged. Lastly, to enlarge the allowed flight region, a real-time ionospheric threat adaptation algorithm was developed to lower the conservatism on VPLs against a conservative ionospheric error model. This study is a stepping stone for research on more-detailed KF-based multi-sensor integrity systems. This study would be extended to assure the navigation integrity of different types of multi-sensor systems for UAS applications as well as other types of unmanned vehicles.