|
The
First EGNOS Trials at Sea:
IN COLUMBUS' WAKE
LUIGI SINAPI,
Istituto
Idrografico della Marina, Genoa, Italy SALLY
BASKER, European GNSS Secretariat, Brussels, Belgium GIORGIO SOLARI,
European Space Agency, Brussels, Belgium HUGUES SECRETAN, European
Space Agency, Toulouse, France JEAN-PIERRE BARBOUX, DSNP, Nantes,
France
The European Tripartite Group is developing the
European Geostationary Navigation Overlay Service (EGNOS), which will
become operational in 2003. Designed as a satellite-based augmentation
to GPS and GLONASS, EGNOS will significantly improve the performance available
from these two systems, facilitating precision and safety applications
for a wide range of users.
Genoa, Italy, one of whose most famous sons was 15th
century explorer Christopher Columbus, recently served as the venue of
a 20th century innovation in navigation: the first marine trials of the
European Geostationary Navigation Overlay Service or EGNOS.
The port city on the Italian Riviera is the base for
the Istituto Idrografico della Marina (IIM), an organisation responsible
for hydrographic surveying in Italian territorial waters, investigating
navigation systems for the Italian Navy, and other tasks. The use of EGNOS
interests IIM as a possible cost-effective alternative to commercial and
local area differential services.
In February 2000 the IIM, in conjunction with the European
GNSS Secretariat (EGS), the European Space Agency (ESA), and DSNP, conducted
the first EGNOS maritime trials in Genoa, using signals from the EGNOS
System Test Bed (ESTB). The institute's ship, the Mirto, was equipped
with four different satellite systems: GPS, Local Area Differential (LAD)
GPS, EGNOS, and Long Range Kinematic (LRK) GPS. It was also equipped with
a total station--consisting of a theodolite combined with electronic distance
measurement equipment--and hydrographic data acquisition systems. The
objectives of these trials were: an assessment of EGNOS accuracy in static
and dynamic modes; a comparison of the EGNOS hydrographic survey of an
area with its 3-D bottom model; the simulation of port entry on an official
IIM electronic nautical chart; and an assessment of the cost-effectiveness
of using EGNOS.
This paper outlines the scope of the trials, presents
the results, and provides an assessment of the use of EGNOS for maritime
operations.
Trials Systems
The EGNOS System Test Bed (ESTB) has been developed as
part of the EGNOS Advanced Operational Capability (AOC) contract. It is
designed as a flexible and responsive system, capable of being expanded
through the deployment of additional Reference and Integrity Monitors
(RIMS), Navigation Land Earth Stations (NLES), or updated Central Processing
Facility (CPF) software to support trials requirements in Europe or elsewhere.
It has a number of roles within the EGNOS development including:
- providing off-air data from the RIMS to the industrial
team developing EGNOS,
- testing and tuning CPF algorithms, and
- broadcasting a representative Signal-in-Space (SIS)
for user trials.
During the trials period the ESTB comprised eight RIMS
distributed around Europe, a CPF at Hönefoss, Norway, and a Navigation
Land Earth Station (NLES) in Toulouse, France (Figure 1). Additional EURIDIS
reference stations in Kourou, French Guyana, Hartebeeshoek, South Africa,
and Toulouse were used to determine the orbit and clock parameters for
the navigation payload on the Atlantic Ocean Region East (AOR-E) satellite
at 15.5ºW.
The RIMS are equipped with external oscillators and dual
frequency GPS/GLONASS/GEO receivers to track the three satellite constellations.
The tracked data are then transmitted to the CPF where they are used to
produce the so-called EGNOS data products --Wide Area Differential (WAD)
corrections and integrity information for each satellite. The data products
are then transmitted to the Navigation Land Earth Stations (NLES) where
they are scheduled to meet time-to-alarm requirements, packed into the
standard message form, modulated on a GPS look-alike signal, and uplinked
to the EGNOS GEOs before being broadcast to the users.
The ESTB was declared operational in February 2000, and
provided a signal-in-space between 08:00 and 17:00 on each day of the
trials.
Navigation Sensors and Systems. The satellite
navigation equipment available for the trials included a dual-frequency
LRK GPS unit to provide ground truth, a single-frequency differential
GPS receiver, a single-frequency EGNOS receiver, and the IIM's two LAD
receivers. The IIM also provided a total station and a ring with 18 reflecting
prisms as an additional source of ground truth, a hydrographic data acquisition
system, a single beam echosounder, and an Electronic Navigation Chart
(ENC) for Genoa.
Platform the Mirto.The trials were
carried out at the IIM and on board the Mirto, which is 44.1m long,
8.5m wide, and has a displacement of 405 tonnes. It has a maximum speed
of 10 knots, and is equipped with instruments for hydrographic and oceanographic
operations.
Trials and Analyses
The trials team performed a static positioning trial
on 17 February 2000, in order to characterise the performance of the EGNOS
and NGPS systems prior to the dynamic trials. The equipment setup is illustrated
in Figure 2.
The two receivers were connected to a single antenna
that was placed on a WGS84 point at the IIM. Data recorded from the EGNOS
receiver were processed, and the periods of bad clock and ionospheric
corrections were cancelled (that is, limited to the periods of non-differential
recording). Processed data were used to extract the EGNOS and NGPS scatter
plot around the geodetic point (Figure 3 and 4).
The EGNOS scatter plot is well distributed except for
three little groups of points shifted from the main mass. These are due
to a degradation of corrections. On the whole the results are quite good
and represent a good starting point for improving and refining the EGNOS
static and dynamic performance. The NGPS results are driven by selective
availability (SA), giving mean and standard deviation for latitude and
longitude of -2.93m 6 25.71m and 3.55m 6 19.54m respectively. The equivalent
EGNOS results for latitude and longitude are 0.36m60.90m and 0.96m61.97m.
Sea-wall Dynamic Trial. A sea-wall dynamic trial
was undertaken on 18 February, during which the Mirto executed
courses parallel to the external breakwater in the harbour of Genoa. The
aim of the trial was to assess the performance of the different systems
in a dynamic environment. The equipment setup is illustrated in Figure
5. Data from the four satellite systems were recorded using the MHYDROS
software. The LRK solution was accepted as ground truth once it was shown
that its accuracy was equivalent to that of the total station.
The dynamic trials route is illustrated in Figure 6.
The LRK, DGPS, and EGNOS tracks overlap, and SA dither causes the NGPS
track to depart from the others. The statistics evaluated are NGPS-LRK,
DGPS-LRK, and EGNOS-LRK. Subtracting LRK effectively gets rid of the dynamic
motion, enabling the performance of the different systems to be assessed.
The differences between EGNOS and LRK for latitude and
longitude are 21.27m61.41m and 20.14m60.75m respectively. These compare
favourably with the LAD equivalent values of 20.58m61.44m and 0.21m60.98m.
Hydrographic Survey. The IIM is the official Italian
cartographic agency. As such, it is responsible for the survey of the
Italian seas and for the production and updating of nautical charts and
publications. It has to comply with the International Hydrographic Organisation's
minimum standards for equipment usage and survey methods.
The area chosen for the trial (Figure 7) presents the
following peculiar features:
The area is divided into two different zones: Zone 1,
up to the 100-metre bathymetric line, is flat and Zone 2 presents a strong
gradient up to the 700-metre bathymetric line (it belongs to one of the
Ligurian canyons).
Zone 1 can be classified, in accordance with Special
Publication n°44 -- IHO "Standards for Hydrographic Surveys" -- as
an area of Order 1 (harbours, harbour approach channels, recommended tracks,
and some coastal areas with depths up to 100 metres) where a horizontal
accuracy (95% of confidence level) of 5 m 1 5% of depth is required.
Zone 2 can be classified as an area of Order 2 (generally
off-shore areas), where an horizontal accuracy of 20 m 1 5% of depth is
required.
The survey scale was 1:25,000.
The choice of an area with different gradients was necessary
to verify the capability of EGNOS to faithfully represent the sea bottom
using 3-D modelling without altering the real shape of the sea bottom
and the effective sounding positions. The horizontal accuracy and the
reliability of the positioning system are extremely important for two
reasons:
- The accuracy of the position of a sounding is the
accuracy at the position of the sounding on the bottom located within
a geodetic reference frame, and for an Order 2 survey using a single-beam
echo sounder, the accuracy of the position is that of the sounding system
sensor.
- An error of echo sounder position directly affects
the insonified area of the sea bottom, and it may depend on the width
of the echo sounder lobe and on the bottom gradient.
Three satellite systems (DGPS, EGNOS and LRK) were connected
with the same single-beam echo sounder through a splitter, as shown in
Figure 8.
The survey was carried out by lines perpendicular to
the bathymetric (contour) lines and at a parallel constant distance of
250 metres from each other. The vessel speed throughout the survey was
about 8 knots. The lever-arm offset between the GPS antennas and echo
sounder transducer was applied in the hydrographic acquisition software
to compute correctly the position of sea bottom images and sounding profiles.
The results of the survey are in the form of trend lines,
graphs with bathymetric lines, and 3-D models. The sea bottom shape and
systems accuracy have been analysed by drawing the bathymetric lines every
10 metres up to the 100-metre bathymetric line, and every 50 metres from
the 100-metre bathymetric line to the 700-metre bathymetric line. The
shifts between these lines have been measured, and the system trends have
been reproduced to locate minimum, medium, and maximum depths.
The trend lines (see Figure 9) show the capability of
the systems to reproduce minimum depths, medium depths, and maximum depths.
An imperfect reproduction of the maximum values of the bottom can be observed,
because depth selection requires, for safety navigation problems, special
attention to minimum depths.
For each of the three pairs of systems (DGPS-EGNOS, LRK-EGNOS,
and LRK-DGPS) 3-D model comparisons were prepared, although only the DGPS-EGNOS
comparison is presented in this paper (Figure 10). There is scarcely any
overlap between the zones produced by the two systems. The reconstruction
of the bottom shape is very accurate. In fact, the positions of the echo
sounder transducer are so close to each other that their differences do
not affect the insonified sea-bottom area.
Finally, an examination of the bathymetric lines (Figure
11) shows that the difference in metres related to the EGNOS system is
in accordance with the survey minimum standards in terms of horizontal
accuracy. In fact the differences are perfectly in line with the horizontal
accuracy requested for Order 1 and 2 surveys, and the greater distances
between LRK and EGNOS bathymetric lines are due to the automatic depth
choices made by the hydrographic evaluation software.
The above pictures and analyses prove that EGNOS can
be used in Order 1, 2, and 3 hydrographic surveys.
Port Approach/Departure. The principal aim of
nautical cartography and more in general of nautical documentation is
to ensure safety in navigation and to help safeguard human life at sea.
This is the primary activity of IIM in Italy.
The International Hydrographic Organisation (IHO), in
conjunction with the International Maritime Organisation, has been working
since 1996 to regulate the development of electronic nautical cartography.
Therefore, a port approach was performed with the aim
of verifying EGNOS' compatibility with the IIM ENC for Genoa. A number
of approach routes were planned on Electronic Navigation Chart n°55
INT 3362 (prepared by the IIM) at scale 1:10,000, and the helmsman steered
the ship on the routes using the MHYDROS software.
The "dual reckoning" function was used to display the
difference between two independent positioning systems in terms of range
(nautical miles) and bearing (degrees), and it also provides positional
accuracy. The Electronic Chart Display Information System (ECDIS) acted
as a "black box," recording time, position, route, speed, depth, and ENC
information at one-minute intervals. The results of the trial have been
obtained by inspection, comparing the on-screen position differences between
the LAD and EGNOS solutions. The average difference was very similar to
the sea wall trial: around 2 metres depending on the data latency. These
results indicated that EGNOS is appropriate for port approach or departure
manoeuvres.
Summary and Conclusions
These sea trials have assessed the potential performance
of EGNOS for the following operations:
- hydrography,
- coastal and precision navigation,
- port entrance and exit operations, and
- use of electronic navigation support (ENC).
- EGNOS system met the hydrographic and maritime requirements
during each trial.
- offers the following attractive features for hydrographic
operations:
- Accuracy: EGNOS is accurate enough for ordinary hydrographic
surveys (that is, excluding harbours, berthing areas and associated
channels with minimum under-keel clearances).
- Versatility: the accuracy of EGNOS (suitable for Order
1 large scale surveys) means that hydrographers do not need to establish
geodetic points on land, thus minimising the logistic support required.
- Signal coverage: The use of geostationary satellites
for broadcasting the EGNOS corrections overcomes the range limitations
of the VHF or UHF communications links used by many DGPS and LRK systems.
- EGNOS has demonstrated its suitability for both coastal
navigation and coastal approach operations. The availability of EGNOS
receivers at prices similar to those of current GPS receivers will encourage
its acceptance as an alternative positioning system in the Mediterranean
Sea. c
Manufacturers
Equipment used in these tests included an Atlas KruppDeso
20 single beam echo sounder, a Trimble 4000 GPS receiver,
a Leica Mhydros hydrographic acquisition system, a Leica
TCA 1101 total station, and four DSNPAquarius 5000 Series
receivers configured in three different ways: dual-frequency LRK, single-frequency
differential GPS, and single-frequency EGNOS.
|
