Session A4: Autonomous Applications

High-Accuracy and Resilient GNSS Receiver for an Autonomous Vehicle
Filipe Carvalho, Ricardo Prata, Bruno Cardeira, Carlota Cardoso, Rui Nunes, Antonio Fernández, Deimos Engenharia
Date/Time: Thursday, Sep. 19, 4:23 p.m.

The general trend of geodata acquisition is that it will be made using robots [1], be them for indoor, aerial (corridor or large-area) mapping, or environmental monitoring [2]. Most arguably, mapping robots will dominate the landscape – or, at least, play a significant role – of geodata acquisition for aerial mapping [3].
To operate safely, LoA 4 (Level of Automation 4) AVs (Autonomous Vehicles) require purpose-built maps – so-called HD (High Definition) maps – which are significantly more detailed, accurate and reliable than conventional navigation maps. These maps, for AVs or other fields, are built upon highly accurate base maps by adding specific geoinformation and later updating these base maps. As opposed to current brute-force common practices where base maps, compiled from time to time, are continuously updated with crowdsourced data, often of questionable quality and positional accuracy (about 1 m), HD maps should be cm level accurate and updated on a daily to weekly basis. Getting the “appropriate” LoA 4 maps is, today, one of the challenges of the automotive industry.
To achieve the desired positional accuracy, the development of highly accurate GNSS receivers is essential. Furthermore, for AV applications, resilience to spoofing attacks, which consist on the generation and transmission of false GNSS signals aimed at misleading the navigation solutions of receivers, is a necessity.
In this context, Deimos Engenharia is developing a GNSS receiver in the scope of the GAMMS (Galileo/GNSS-Based Autonomous Mobile Mapping System) project, an EU funded Horizon2020 project. This project aims at developing an AMMS (Autonomous terrestrial Mobile Mapping System), based on the tight integration of AV, navigation/geodetic, and AI (Artificial Intelligence) technologies. The GAMMS consortium comprises different Industry and academic partners, namely Deimos Engenharia, École Polytechnique Fédérale de Lausanne, ENIDE, GeoNumerics, GEOSAT, Pildo Labs, Sensible4 and Solid Potato.
The navigation system of GAMMS includes the Galileo-based GNSS receiver that is the subject of this paper, IMUs (Inertial Measurement Units), LiDAR (Light Detection and Ranging) and other sensors and a trajectory determination software module, which will receive information from the receiver and the sensors to compute a highly accurate trajectory of the autonomous vehicle.
To obtain high accuracy positioning and resiliency to spoofing attacks, the GNSS receiver for GAMMS will feature state-of-the-art Galileo features, such as Galileo OSNMA (Open Service Navigation Message Authentication) [4] and HAS (High Accuracy Service) [5]. The receiver will also make use of Galileo E5 AltBOC [6] signals to improve its performance in urban areas, where multipath issues are more prevalent. All these Galileo features and services will be tested in live conditions and their effects measured on the performance of the receiver and navigation solution. This paper will focus on the design and implementation of OSNMA and HAS services on the receiver, two key features of the project, and the results obtained by these features during testing activities. These consist on two test campaigns, where a demonstrator and later a prototype of the GAMMS solution will be tested. Further unit testing of the receiver will be taking place in Deimos’ facilities, with the use of a GNSS RFCS (Radio Frequency Constellation Simulator) and a GNSS antenna.
The GNSS receiver architecture is based on a Software Defined Receiver (SDR), composed of a triple frequency RF-FE (Radio Frequency – Front End) and a SoM (System-on-Module) that integrates a Xilinx/AMD Zynq UltraScale+ MPSoC (Multi-Processor System-on-Chip). During operation, the RF-FE unit processes E1, E5 and E6 signals, feeding the DSP (Digital Signal Processing) chain and the respective receiver real-time software, which are running on the MPSoC. The proposed solution allows the generation of dual frequency raw measurements using Galileo E1 and E5 bands, the retrieval of navigation data including OSNMA bits, and the allocation of E6 data channels for the reception of the HAS corrections.
Galileo OSNMA is a service that allows receivers to authenticate the signals received, improving their resilience to spoofing attacks. This service broadcasts OSNMA information through the E1 I/NAV navigation message, using 40 bits of each odd Galileo I/NAV page, previously reserved. The receiver will process these bits and authenticate the navigation data of the satellites in view [7]. Satellites not broadcasting OSNMA may still be authenticated via cross-authentication features of the OSNMA service. If the receiver detects spoofing of a signal, it will inform the user, and the GNSS measurements and navigation data of the SVID broadcasting that signal will be rejected.
Special features of OSNMA will also be implemented, including TESLA (Timed Efficient Stream Loss Tolerant Authentication) Chain renewals and revocations, Public Key renewals and revocations and OSNMA Alert Messages. Some live tests have been performed on a preliminary implementation of OSNMA, where the receiver was able to obtain and process OSNMA bits and authenticate messages, as well as enact a TESLA Chain change (e.g. on a Live Test performed by GSC (European GNSS Service Centre) on December 1st 2023 at 14:00 UTC).
Galileo HAS is a service that aims at providing corrections to improve ephemeris information, pseudorange and carrier phase estimates and atmospheric impacts. It is transmitted through E6-B signal in Galileo C/NAV pages, which are broadcast with a period of 1 second. The receiver developed for GAMMS acquires and tracks E6-B signals and process the C/NAV pages, namely the HAS component of each page, or HAS page. HAS messages are composed of several HAS pages, with each message having a maximum possible size of 32 pages and a minimum of 1, unlike other messages, such as I/NAV or F/NAV, which are composed of a constant number of pages.
A FEC (Forward Error Correction) decoder for E6-B C/NAV pages is implemented, using the same convolutional decoding scheme as the one used for other Galileo data signals. The main differences between HAS messages and other messages are: different satellites broadcasting HAS contribute to the same HAS message, parallelizing its distribution; HAS messages are further encoded through an outer layer scheme – HPVRS (High Parity Vertical Reed-Solomon) – which improves the reception efficiency of HAS messages [8].
In GAMMS, the receiver will collect the E6 C/NAV pages, reconstruct the full HAS message by obtaining the proper number of encoded pages and Reed-Solomon decoding them, and convert the data obtained into useful parameters – in both a custom and RTCM (Radio Technical Commission for Maritime Services) formats) – to feed the GAMMS trajectory determination software unit.
[1] Thrun,S., “Robotic mapping: a survey”, 2002, School of Computer Science, Carnegie Mellon, University, Pittsburgh, PA, USA, pp. 31.
[2] Bayat,B., Crasta,N., Crespi,A., Pascoal,A.M., Ijspeert, A., “Environmental monitoring using autonomous vehicles: a survey of recent searching techniques”, 2017, Current Opinion in Biotechnology, Vol. 45, pp. 76–84. doi :10.1016/j.copbio.2017.01.009.
[3] Patias,P., Armenakis,C., “Unmanned Vehicle Systems in geomatics: towards robotic mapping”, 2019, Whittles Publishing, Dunbeath, Caithness, Scotland, UK, pp. 320.
[4] European Union Agency for the Space Programme, “Galileo Open Service Navigation Message Authentication (OSNMA) Info Note”, 2021.
[5] European Union Agency for the Space Programme, “Galileo High Accuracy Service (HAS) Info Note”, 2020.
[6] European Commission, “Galileo Open Service Service Definition Document (OS SDD)”, Issue 1.3, November 2023.
[7] European Commission, “Galileo Open Service Navigation Message Authentication (OSNMA) Signal-In-Space Interface Control Document (SIS ICD)”, Issue 1.1, October 2023.
[8] European Commission, “Galileo High Accuracy Service Signal-In-Space Interface Control Document (HAS SIS ICD)”, Issue 1.0, May 2022.