Previous Abstract Return to Session D3 Next Abstract

Session D3: GNSS Augmentation and Robustness for Autonomous Navigation

Design and Evaluation of a GNSS/INS Aircraft-based Augmentation System
O. Garcia Crespillo, German Aerospace Center (DLR); M. Joerger, Virginia Tech; J. Skaloud, Swiss Federal Institute of Technology (EPFL); M. Meurer, DLR
Date/Time: Thursday, Sep. 22, 10:40 a.m.

Safety-critical navigation applications require that estimation errors be reliably quantified. Over the last two decades, significant effort was spent towards guaranteeing bounds on Global Navigation Satellite Systems (GNSS)-based position estimation errors together with satellite-based or ground-based augmentation systems (SBAS, GBAS) and Advanced Receiver Autonomous Integrity Monitoring (ARAIM) for aviation application. This was achieved by careful modeling of GNSS ranging measurement errors (among others, satellite clock and orbit errors, tropospheric and multipath delays) and by rigorous algorithm design that quantify integrity and continuity risks in all circumstances, including in the presence of undetected faults.
In parallel, in Aircraft-based Augmentation Systems (ABAS), an Inertial Reference System (IRS) has been traditionally employed as an additional source of redundant navigation information. Recently, with improved capabilities and tighter integration scheme with single-frequency GNSS, the reliance on Inertial Navigation Systems (INS) in certifiable safety-critical GNSS-based systems has increased as captured in Minimum Operational Performance Standard (MOPS) [1]. However, further modeling and algorithm developments are needed to incorporate GNSS/INS in future dual-frequency, multi-constellation GNSS standards.
Safety-critical GNSS methods like the ones used in ARAIM use “snapshot” error models and algorithms: position estimation is performed independently at every time epoch, regardless of past-time estimates. In contrast, GNSS/INS systems are typically performed using a Kalman filter (KF). This has the following implications for the evaluation of integrity, availability, and continuity of GNSS/INS systems:
- First, measurement error modelling and estimation must consider the time-correlated nature of GNSS and INS errors. Particularly, new overbounding methods for time-correlated errors must be implemented in order to guarantee safe KF error estimation.
- Second, these error models must be incorporated in the KF for proper propagation of the individual error bounds. This includes accounting for nominal bounded biases due to signal deformation affecting code measurements, which are modeled differently in ARAIM than satellite clock and orbit ephemeris, troposphere, and multipath error and receiver noise.
- Third, GNSS/INS performance depends on aircraft trajectory profile and on KF initialization and current state convergence, including uncertainty of the attitude and inertial biases. Therefore, assumptions are needed to carry out ARAIM availability and continuity analyses.
In the first part of this paper, we present a tight GNSS/INS integration scheme that uses Multiple Hypothesis Solution Separation (MHSS) for fault detection and integrity monitoring. This approach builds upon the latest developments of the EU-US WG-C’s algorithms, error models, and Integrity Support Message (ISM) parameters [2]. Suggestions are made on algorithm, model, and ISM extensions for sequential measurement processing. We describe recursive models for time-correlated residual code and carrier-phase errors. We incorporate these models in the KF by state augmentation to ensure that nominal navigation errors are properly overbounded [3,4]. This design is compared to current single frequency GPS/INS design from MOPS [1].
In the second part of the paper, we propose a new methodology to analyze availability and continuity using specific approach and landing procedures in place at worldwide airport locations. These realistic testing profiles enable the availability and continuity evaluation of a “sequential” GNSS/INS MHSS algorithm and a fair comparison with ARAIM during aircraft maneuvers.
The third part of the paper aims at validating the algorithms through (1) simulations over the GPS and Galileo constellation repeatability periods (2) processing of real flight data collected by the German Aerospace Center (DLR) using the Falcon E20 aircraft.
The performed evaluations show the potential of GNSS/INS to achieve tighter requirements than “snapshot” ARAIM during realistic aircraft maneuvers. In particular, our GNSS/INS ARAIM method guarantees safe navigation in the presence of time-correlated errors; this is achieved while maintaining availability and continuity despite losing satellite visibility due to aircraft banking.
[1] RTCA, “DO-384 Minimum operational performance standards (MOPS) for GNSS aided inertial systems”, Radio Technical Commission for Aeronautics, Tech. Rep. DO384, 2020
[2] Working Group C - ARAIM Subgroup, “Milestone 3 Report”, EU/US Cooperation on Satellite Navigation, Tech. Rep., 2016
[3] O. Garcia Crespillo, M. Joerger, and S. Langel, “Overbounding GNSS/INS integration with uncertain GNSS Gauss-Markov error parameters”, in Position, Navigation and Timing Symposium (PLANS), May 2020
[4] O. Garcia Crespillo O, “GNSS/INS Kalman Filter Integrity Monitoring with Uncertain Time Correlated Error Processes”. PhD Thesis, EPFL; 2022.



Previous Abstract Return to Session D3 Next Abstract