High-Integrity Modeling of Nonstationary Noise Processes for GNSS/INS Integration

Abstract

This paper describes a power spectral density (PSD) bounding method for deriving high-integrity models of stationary and nonstationary time-correlated measurement error processes. The method is intended for safety-critical commercial aircraft navigation where robust sensor models are required to predict error bounds on position and orientation estimates. These bounds are used both in navigation system design for integrity performance analyses and in operation to determine whether a pilot should proceed with an operation. In prior work, we used PSD upper-bounding to obtain high-integrity models of time-correlated global navigation satellite system (GNSS) measurement errors. However, the method was limited to stationary processes. In this paper, we derive an approach to expand the concept of PSD bounding to nonstationary error modeling for Kalman filter-based estimation using GNSSs and inertial navigation systems in aircraft navigation applications.