=Paper=
{{Paper
|id=Vol-2880/paper16
|storemode=property
|title=State of the Art of the Agriculture Professional GNSS Receivers
|pdfUrl=https://ceur-ws.org/Vol-2880/paper16.pdf
|volume=Vol-2880
|authors=Jacopo Capolicchio,Daniele Mennuti,Ileana Milani,Luigi Villa,Joaquin Reyes Gonzalez,Martin Sunkevic
|dblpUrl=https://dblp.org/rec/conf/icl-gnss/CapolicchioMMVG21
}}
==State of the Art of the Agriculture Professional GNSS Receivers==
https://ceur-ws.org/Vol-2880/paper16.pdf
State of the Art of the Agriculture Professional GNSS Receivers
Jacopo Capolicchioa, Daniele Mennutib, Ileana Milanic, Luigi Villad, Joaquin Reyes Gonzaleze
and Martin Sunkevice
a
Thales Alenia Space Italia, Via Saccomuro 24, 00131 Rome, Italy
b
Business Integration Partners, Via Sicilia 43, 00187 Rome, Italy
c
Randstad Italia, Via Tiburtina 1072, 00156 Rome, Italy
d
Akka Italia, Corso Enrico Tazzoli 215/12/B, 10137 Torino, Italy
e
European GNSS Agency (GSA), Janovského 438/2, 170 00 Prague 7, Holesovice Czech Republic
Abstract
The aim of this paper is to present the outcomes of the agriculture testing campaign performed
by Thales Alenia Space Italia in the implementation of a contract signed with the European
GNSS Agency (GSA) and financed by the European Union under the Galileo Programme
budget. The main objective of the test campaign is to evaluate the performance of a set of
professional GNSS receivers, highlighting the added value of using the Galileo system in the
GNSS market segment of agriculture Machine Guidance, in particular from the end user point
of view. This paper will present anonymized performance of eight agriculture receivers, tested
in parallel under the same live conditions, considering different configurations and
augmentation modes. More specifically, GNSS Single Point Positioning (with both single
frequency (SF) and multi-frequency (MF) approach), Satellite Based Augmentation System,
Precise Point Positioning and Real Time Kinematic modes have been tested with single-
constellation and multi-constellation (MC) configurations, considering GPS, Galileo and
GLONASS. The most relevant Key Performance Indicators (KPIs) for agriculture-related
applications, such as cross-track accuracy and repeatability have been assessed per each test
case. The results have shown that Galileo standalone configuration provides similar or even
better performance than GPS standalone, despite the lower number of available satellites with
respect to GPS. Moreover, it is also confirmed that its use in a multi-constellation
configuration, especially for standalone positioning, enhances the performance for both
positioning accuracy and availability.
Keywords 1
GNSS, Galileo, Agriculture Market Segment.
1. Introduction
In recent years, an increasing number of applications started to rely on Global Navigation Satellite
Systems (GNSS) to provide improved services to their users, thanks to the possibility to get accurate
Position, Navigation and Timing (PNT) solutions with cost-effective devices.
Among the other GNSSs, the satellites and ground infrastructure of the European navigation system
Galileo were declared operationally ready on December 2016. In that moment, the Galileo Initial
Services started to be offered worldwide. The performances and limitations of the Galileo Initial
Services, together with the system configuration, are described in [1]. Galileo offers a highly accurate
service but, for the time being, the system is not yet in Full Operational Capability (FOC), therefore the
performance will further increase when the full constellation will be available. Information related to
the Galileo satellites available for Position, Velocity and Time (PVT) computation can be found in [2].
ICL-GNSS 2021 WiP Proceedings, June 01–03, 2021, Tampere, Finland
EMAIL: jacopo.capolicchio@thalesaleniaspace.com (A. 1); daniele.mennuti@mail-bip.com (A. 2); ileana.milani-
somministrato@thalesaleniaspace.com (A. 3); luigi.villa@akka.eu (A. 4); joaquin.reyes@gsa.europa.eu (A. 5);
martin.sunkevic@gsa.europa.eu (A. 6);
©️ 2021 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org) Proceedings
In the last years, encouraged especially by the benefits provided by the use of multi-constellation
systems, GNSS receiver manufacturers have started to provide to their users the possibility to exploit
also Galileo signals. For this reason, the European GNSS Agency (GSA) decided to carry out dedicated
testing campaigns to check the status of the Galileo implementation in professional receivers. In 2018,
a first testing campaign was performed to support the Geographic Information System (GIS)
community. Interesting results have been derived from that, considering also the limited number of
satellites available for the PVT computation, [3]. Due to these encouraging results and the increasing
interest of Precision Agriculture on GNSS-enabled solutions, on July 2020, a second campaign has been
carried out, this time addressing the agriculture market segment. The data collection has been performed
with a number of twenty-two active Galileo satellites.
In order to provide information regarding the state of the art of the GNSS in the agriculture market
segment, in this work the performance of eight professional GNSS receivers has been evaluated and
compared.
Similarly to the testing activities performed in [3], the main objectives of this campaign were:
1. to support the professional receiver’s manufacturers in fine-tuning the implementation of
Galileo within their receivers and pointing out the benefits on its use;
2. to highlight the added value of using the Galileo system in the GNSS market segment of
agriculture Machine Guidance, in particular from the end user point of view;
3. to assess the added value of Galileo in multi-constellation receivers, including an assessment
of where today Galileo is with respect to the other GNSSs.
It is important to remark that the scope of this activity was not the validation of the Galileo system,
but the assessment of the benefits on using Galileo by end users whom will use agriculture receivers
currently available in the market.
Since each agriculture application requires a specific level of accuracy, as described in [4], different
positioning modes have been tested in this testing campaign:
Single Point Positioning (SPP, also referred to as Standalone Positioning);
Satellite Based Augmentation System (SBAS);
Precise Point Positioning (PPP);
Real Time Kinematic (RTK).
In addition, multi-frequency and multi-constellation configurations have been also tested, in order
to evaluate the increase of the accuracy and availability of the PVT solution. In particular, for MF test-
cases all the available frequencies supported by the receivers have been enabled for the PVT estimation:
GPS: L1/L2/L5;
Galileo: E1/E5a/E5b/E5AltBOC
GLONASS: L1/L2/L3
The paper is organized as follows. In Section 2, the set-up of the testing campaign has been described
together with the main KPIs relevant for the agriculture market segment. The main results of the testing
campaign have been presented in Section 3, through the comparison of the performance of the tested
receiver, the demonstration of the Galileo added values and the summary of the agriculture receiver
status. Finally, the conclusions of this work are drawn in Section 4.
2. Agriculture test campaign set-up and KPIs
The test set-up has been organized in a way so that only the performance of the GNSS receivers
were tested and evaluated. IMU and other sensors or also processing on machine guidance
device/software side thus have not been considered.
One of the main objectives of the testing campaign is to assess the added value of Galileo in multi-
constellation receivers, including an assessment of where today Galileo is with respect to the other
GNSS, in particular GPS and GLONASS. The testing campaign envisaged different test-cases in Open
Sky conditions, which is the most relevant scenario for agriculture market segment, with the duration
of three hours per each, testing different configurations and augmentation services.
To obtain comparable results under the same conditions and to ensure the repeatability of the test as
well as an easy controllability of the same, a rail track and a carriage were used to perform the test,
instead of real tractors able to be automatically driven by a configurable GNSS receiver. In particular,
tests were performed in parallel by connecting all the Receivers Under Test (RUTs) to the carriage on
the rail track. During each test-case, the carriage performed several working lines, from point A to point
B, as shown in Figure 1.
Figure 1 Testing Campaign Working Line
After reaching point B, the carriage was driven on reverse mode up to the point A, in the following
line, stop and then continue forwards the rail track towards to the point B. This job lasts approximately
2 minutes, depending on the carriage speed (nominally 7 km/h). It shall be noticed that the rail track
ensures a smooth trajectory, without the typical deviations and vibrations of real agriculture
applications, due to the field irregularities.
For each test case, the following agriculture-related KPIs have been assessed:
Trajectory Error is the variation between the actual tilling trajectory with respect to the
reference one. As it is often used in Precision Agriculture applications, the trajectory error will be
provided in terms of cross-track accuracy.
Repeatability (or year-to-year) is generally understood to mean the ability of a GNSS receiver
or GNSS guidance system to bring the user back to the exact same spot in the field reliably each
time the tractor drives into the field, from year to year. It is worth mentioning that ‘year-to-year’ is
a term used in agriculture community and it is generally used to mean absolute accuracy.
In Figure 2, the methodologies for the calculation of these KPIs are presented, having considered
also [6] and [7]. In particular, the cross-track error has been calculated as the difference between the
instantaneous estimated position with respect to the “true trajectory” in the transversal direction to the
corresponding rail track, while for this testing campaign the repeatability has been calculated as the 95th
percentile of the absolute horizontal accuracy provided by the receiver on two fixed points, indicated
as A and B in Figure 2. With regards to the latter point, the reader should interpret the repeatability as
absolute horizontal accuracy for comparison purposes with other testing campaigns.
Figure 2: Agriculture-related KPIs: cross-track error (on the left) and repeatability (on the right).
Six RUTs over eight are smart antennas, while the other two receivers were connected to a single
shared external antenna. It is worth mentioning that the Test Site was equipped also with a Single Base
Station, able to stream RTCM (Radio Technical Commission for Maritime Services) corrections to the
RUTs, in order to perform RTK test cases.
For each test case, the reference “true” trajectory has been estimated by exploiting a commercial
third-party post-processing tool, that is able to provide kinematics solution by using the rover and base
station’s observables, reaching cm-level 3D accuracy. This is used to benchmark RUTs performance
by using an external tool that is not dependent on the receivers’ proprietary algorithms.
The performance has been evaluated with real-time positions computed by the receivers and,
consequently, by using NMEA (National Marine Electronics Association) messages.
Considering the main objectives of the testing campaign, a test duration of 3 hours is a fair trade-off
between the high number of test cases to be executed and the quality of the results obtained.
Furthermore, 3 hours of data collection are able to provide a relevant number of samples for
performance estimation, as the scenario of interest is Open Sky.
3. Agriculture test campaign results
The main results of the agriculture testing campaign are derived through the evaluation of the
performance achieved by the tested receivers in terms of cross-track accuracy and repeatability.
3.1. Receivers Performance Comparison
The performance of the tested receivers has been compared considering different positioning modes.
More specifically, as introduced in Section 1, both the configuration without corrections (SPP) and the
results achieved by applying augmentation strategies (SBAS, PPP and RTK) have been analyzed.
The comparison of the performance achieved by the different receivers for the cross-track accuracy
and the repeatability is shown in Figure 3 and Figure 4, respectively, where histograms representing the
95th percentile of the related errors are displayed for each positioning mode.
Figure 3: Cross-track Accuracy Histogram [m].
Figure 4: Repeatability Histogram [m].
A selection of the most relevant results among all the tested configurations is shown in these figures.
In particular, from Figure 3 and Figure 4, it is possible to notice that the exploitation of enhanced
configurations, such as the MC-MF approach, with GPS, Galileo and GLONASS constellations and
all their frequencies enabled, provides a significant improvement in performance in terms of cross-track
accuracy and repeatability with respect to the standalone GPS in SF mode for all the tested receivers.
This is mainly due to the fact that the multi-constellation configurations provide better HDOP, as the
number of satellites available for the PVT computation increases.
The SBAS-aided positioning performances are compared only with GPS SF (L1 band), as at the
time of writing this is the only configuration supported by EGNOS. As expected, the application of
SBAS corrections on GPS L1 noticeably improves the positioning performances over the standalone
GPS SF configuration, reaching cross-track accuracy down to 60 cm for almost all the tested receivers.
For PPP and RTK positioning modes, only MC configurations with all frequencies enabled have
been tested, as they represent the situation closest to the end user's needs.
From Figure 3 and Figure 4, it is apparent that not all the receivers could perform PPP test-cases.
However, all the tested configurations provide excellent performance in line with expectations [5] for
almost all the RUTs that support PPP, thus representing a suitable choice for applications requiring dm-
level accuracy, such as Machine Guidance. On average, triple-constellation configuration provides
better performance, especially in terms of cross-track accuracy, with respect to double-constellation
configurations, except from one receiver that probably does not properly manage the three different
constellations.
Finally, when RTK mode is used, all the receivers achieve cross-track accuracy and repeatability at
cm-level, being able to meet the stringent requirements of the most demanding applications such as
Automatic Steering and Variable Rate Application (VRA), as defined in [4].
3.2. Galileo Added Value
As previously mentioned, one of the objectives of this work is to evaluate the performance of Galileo
and assess it versus other GNSS. The results of this testing campaign have shown that in most of the
tested configurations, Galileo brings an added value in terms of cross-track accuracy and repeatability.
Indeed, although the Galileo constellation was not yet fully deployed, it is recorded that Galileo
provides better positioning accuracy performance than GPS, in terms of cross-track accuracy and
repeatability, reaching sub-meter level in Open-Sky scenario.
An example of results showing the added value of Galileo is reported in Figure 5, where the Galileo
SPP and GPS SPP, both with all frequencies enabled, are compared through the Cumulative
Distribution Functions (CDFs) of the cross-track accuracy (on the left) and the repeatability (on the
right) for one of the tested receivers.
Figure 5: Galileo vs GPS agriculture performance: CDF of cross-track accuracy (left) and CDF of
Repeatability (right).
Aiming at showing the overall comparison of GPS and Galileo in SPP mode, in Figure 6, the
performance achieved by all the considered RUTs are shown for different combinations of frequencies
and constellations. On average, Galileo provides very similar or even better cross-track accuracy with
respect to GPS (despite a higher HDOP due to the low number of satellites). It also provides better
repeatability with respect to GPS for almost all the receivers. Moreover, it is important to underline that
the results have shown an availability always at 100% for all the tested receivers when Galileo signals
are exploited, except from one receiver that is not able to perform PVT when less than 5 usable satellites
are present. On the other hand, it is worth to highlight that some of the tested receivers do not support
Galileo only positioning mode, so a full comparison was not possible for all the receivers, as it is evident
in Figure 6.
The Galileo added value can be also seen in multi-constellation configurations, that is reported in
Figure 6 for comparison. This aspect is apparent also observing the results shown in Figure 3 and Figure
4, where we have already underlined that the exploitation of more GNSSs leads to an improvement in
performance with respect to those achieved by GPS standalone positioning mode, especially thanks to
the better HDOP provided by a higher number of satellites available for the PVT estimation.
SPP Cross-Track Accuracy
2
Cross-track
accuracy
95%ile
1
0
GPS SF GAL SF GPS MF GAL MF GPS GAL GLO
MF
RX1 RX2 RX3 RX4 RX5 RX6 RX7 RX8
SPP Repeatability
3
Repeatability
95%ile
2
1
0
GPS SF GAL SF GPS MF GAL MF GPS GAL GLO
MF
RX1 RX2 RX3 RX4 RX5 RX6 RX7 RX8
Figure 6: SPP results with different combinations of constellations and frequencies: cross-
track accuracy and repeatability
3.3. Agriculture receiver status
In this section, the state of the art of agriculture professional receivers is shown through the
comparison between the performance achieved by the receivers used in this testing campaign and the
accuracy values expected for agriculture applications. More specifically, in Figure 7 the average
performance of all the receivers for all the analyzed positioning modes has been benchmarked with
respect to agriculture user needs and requirements of different applications, which are presented in [4].
By observing this figure, it can be seen at a glance that on average the tested receivers provide
performance in line with expectations for almost all the analyzed positioning modes and configurations,
while the need of improvements has been highlighted for Galileo only standalone mode (both single-
frequency and multi-frequency) set for some RUTs for the first time and only for the purpose of this
particular testing, as it can be noticed from Figure 6 (i.e. Galileo SF for Receiver 1). It has to be noted
that it is highly improbable mode as most of the users want all what is available, that is multi-
constellation mode in which Galileo brings indisputable added value to all RUTs.
Figure 7: Average performance highlighted by the testing campaign. Required cross-track accuracy is
defined in [4].
4. Conclusions
In this paper, the Galileo added value in the agriculture market segment and the assessment of the
status on its implementation in the specific professional receivers have been addressed. The evaluation
has been performed through the analysis of the results obtained on the data acquired in the agriculture
testing campaign held in July 2020 and performed by Thales Alenia Space Italia in the implementation
of a contract signed with the European GNSS Agency (GSA) and financed by the European Union
under the Galileo Programme budget.
The results presented in this paper show that, although the Galileo system is not yet in Full
Operational Capability, with not fully deployed infrastructure (mainly in terms of available satellites)
there are several benefits that the users experience already today. The analyses on the standalone
positioning mode confirms that Galileo standalone configuration provides similar or even better
performance than GPS standalone, despite the lower number of available satellites with respect to GPS.
In addition, it is also confirmed that their joint use brings an increase in performance in terms of both
positioning accuracy and availability. Finally, the results on the test cases with augmentation methods
(SBAS, PPP and RTK) have shown that the performance is in line with expectations for almost all
configurations, confirming that the tested receivers are able to meet also the needs of the most
demanding applications.
5. References
[1] OS Service Definition Document, Issue 1.1 May 2019.
[2] https://www.gsc-europa.eu/system-status/Constellation-Information.
[3] M. Eleuteri, J. Capolicchio, and S. Perugia. “Testing Galileo Performances of GIS Commercial
Receivers in Different Operational Environments.” 24th Ka and Broadband Communications,
Navigation and Earth Observation Conference, October 2018.
[4] European GNSS Agency, Report on Agriculture User Needs and Requirements, Issue 2.0, July
2019.
[5] V.I. Adamchuk, T.S. Stombaugh, and R.R. Price, GNSS-Based Auto-Guidance in Agriculture, Site
Specific Management Guidelines, SSMG-46, August 2008
[6] ISO 12188-1:2010, Tractors and machinery for agriculture and forestry — Test procedures for
positioning and guidance systems in agriculture — Part 1: Dynamic testing of satellite-based
positioning devices
[7] ISO 12188-2:2010, Tractors and machinery for agriculture and forestry — Test procedures for
positioning and guidance systems in agriculture — Part 2: Testing of satellite-based auto-guidance
systems during straight and level travel