A New Method for Multiple Outliers Detection in Receiver Autonomous Integrity Monitoring

Abstract

Integrity of Global Navigation Satellite System (GNSS) includes the ability of a system to provide timely warnings to the user when the system should not be used for the intended operation. Receiver autonomous integrity monitoring (RAIM) is important for safety of life and liability critical applications. Outlier detection and fault exclusion are essential issues for RAIM so that GNSS positioning solutions are not susceptible to errors. Most of existing RAIM algorithms provide adequate integrity for only a single-satellite fault because there is an extremely small probability that significant simultaneous multiple outliers may occur. However, with the development and wide applications of multiple satellites navigation systems with multiple constellations, it is not realistic under assumption of a single outlier once again. RAIM will be required to detect the presence of multiple satellite failures. The study has proposed a new method to select the quasi-accurate observations by L 1 norm method and median and the principle of selecting quasi-accurate observations has been built. Then the quasi-accurate detection of outliers is employed to detect multiple outliers simultaneously. The detail analysis of GNSS single point position example has been conducted to assess the performance of the proposed approach. The results show that the new proposed method is capable of identifying and isolating multiple outliers accurately so that the reliable positioning solution is guaranteed.