Integrity Monitoring for High-Precision Navigation in Multi-Fault Scenarios

Abstract

High-precision positioning is fundamental to ensure safe navigation in the field of autonomous transportation, where Global Navigation Satellite System (GNSS) play a pivotal role. However, navigational demands for Safety-of-Life (SoL) applications go beyond high accuracy, with integrity representing a critical aspect. This involves the integrity monitoring of the GNSS systems to ensure the high demanding requirements in terms of accuracy, integrity, and continuity are met. The use of GNSS carrier phase measurements is fundamental to obtain high-precision positioning solutions such as Precise Point Positioning (PPP), however it entails dealing with the complexity of integer phase ambiguities resolution. Furthermore, there is a lack of integrity concepts today accounting for absolute precise carrier phase positioning with integer ambiguity resolution and able to perform under multiple simultaneous faults, as it is the case for urban environments. This thesis aims at developing a methodological approach to assess the performance and detectability of integrity threats using robust estimators while exploiting carrier phase information. On the one hand, Multiple hypothesis solution separation (MHSS) schemes are well-established methods for integrity monitoring which involve considering different fault hypotheses (or fault modes) a-priori. Thus, it requires to run in parallel a bank of Kalman filters, one for each fault mode considered. This approach suffers from a high computational load and becomes increasingly complex when multiple faults are considered. On the other hand, Robust Statistics-based estimators have shown improved performance and resilience to outliers with respect to classical estimators under nominal and non-nominal conditions (i.e., when the Gaussian assumption is not hold). However, their applicability to integrity monitoring is still a field to be explored. It has been shown that the use of M-estimators for filtering can achieve higher accuracy performance of PPP float-ambiguity solution to a standard Kalman Filter even when outliers exist in measurements. To further explore the ability of robust filtering technique in PPP integrity monitoring, this thesis will focus on applying a robust estimator to achieve PPP fix-ambiguity solution, provide associated integrity information and compare its performance to the conventional MHSS scheme.