International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 5
Number 2
June 2011
149
1 INTRODUCTION
The maritime integrated PNT System (Figure 1) is
the sum of satellite-based, ashore and aboard com-
ponents. The integrated use of these components en-
ables the accurate and reliable provision of position,
navigation and timing data to all maritime applica-
tions.
Position fixing systems are identified as one
strategic key element of e-navigation [1]. Existing
and future Global Navigation Satellite Systems
(GNSS) like GPS, GLONASS and GALILEO are
fundamental infrastructures for global positioning.
Additionally, terrestrial services are used or consid-
ered as candidates to improve the positioning per-
formance (augmentation services: e.g. IALA Beacon
DGNSS, RTK) or to ensure the backup functionality
(backup services: e.g. e-LORAN, R-Mode) respec-
tively to GNSS. Due to their interoperability and
compatibility these systems can be used alternatively
or complementary for positioning, navigation and
timing.
The International Association of Marine Aids to
Navigation and Lighthouse Authorities (IALA) has
introduced the term “integrated PNT device” to de-
scribe the on-board part of maritime, integrated PNT
system. In [2] the integrated PNT device is de-
scribed as “a device using any available IMO recog-
nized radio navigation systems simultaneously to
provide the best electronic position fix for the ship”.
Following this definition, the outlined objective of
the PNT device is focused on the provision of posi-
tion information to different applications. Several
performance standards for shipborne GNSS and
DGNSS receivers were developed and approved by
IMO in the last decade: GPS [3], GLONASS [4],
DGPS and DGLONASS [5], combined
GPS/GLONASS [6], and GALILEO [7]. A logical
consequence of this standardization process could be
the preparation of a new performance standard for a
multi-system radio navigation receiver as core ele-
ment of the on-board part of the PNT System (Fig-
ure 1). A more generally admitted approach can be
achieved by the introduction of the PNT Unit.
The on-board PNT Unit aims at the provision of
position, navigation and timing data in accordance
with specified performance requirements, which
change during berth to berth navigation. The core of
the on-board PNT Unit is a value-added processing
system using available radio navigation systems and
services in combination with on-board sensors for
accurate and reliable PNT-data provision. The on-
board PNT Unit is on the one hand part of the inte-
grated PNT system and on the other hand part of the
on-board Integrated Navigation System (INS).
Concept for an Onboard Integrated PNT Unit
R. Ziebold, Z. Dai, T. Noack & E. Engler
Institute of Communications and Navigation, German Aerospace Centre (DLR), Neustrelitz,
Germany
ABSTRACT: A robust electronic position, navigation and timing system (PNT) is considered as one of the
core elements for the realization of IMO-s (International Maritime Organization) e-Navigation strategy. Ro-
bustness can be interpreted as the capability of an integrated PNT system to provide PNT relevant data with
the desired accuracy, integrity, continuity and availability under consideration of changing application condi-
tions and requirements. Generally an integrated PNT system is a composite of service components – like
GNSS, Augmentation Systems and terrestrial Navigation Systems – and an on-board integrated PNT Unit,
which uses the available navigation and augmentation signals in combination with additional data of sensors
aboard to provide accurate and robust PNT information of the ship. In this paper a concept of such an on-
board integrated PNT Unit will be presented, which is designed to fulfill the specific user requirements for
civil waterway applications. At first, the user requirements for an integrated PNT Unit will be overviewed.
After that, existing integrity monitoring approaches will be analyzed. Finally, a first integration scheme for an
integrated PNT Unit will be presented with a special focus on the internal integrity monitoring concept.
150
Figure 1. Integrated PNT System (dark grey: standard, light
grey: considered options)
Due to user needs such as “Indication and Im-
provement of Reliability” and “Alarm Management”
[1] identified within the framework of the e-
Navigation process, a more general approach for the
on-board part of the integrated PNT system shall be
aimed. Reasons for this perspective are on the
one hand the need for redundancy to improve the ro-
bustness of PNT-information and to enable the as-
sessment of accuracy by suitable integrity monitor-
ing functions. On the other hand the type of
implementable redundancy (equipment, different
measurement methods, over determined systems, al-
ternative applicable techniques) specifies the poten-
tial of error detection, identification and mitigation.
2 TECHNICAL REQUIREMENTS AND
SENSORS
2.1 Technical requirements
In [8] the aim of GNSS is described as a system to
provide worldwide position, velocity and time de-
termination for multimodal use. An operational re-
quirement indicates the tasks of shipborne GNSS
devices in the provision of position, time, course and
speed over ground. But in Appendix 2 and 3 of the
same document minimum maritime user require-
ments are only given for horizontal position. Other-
wise the use of the item Electronic Position Fixing
System (EPFS) in [9] implicates that the scope of
GNSS is rather the provision of position data than
the provision of PNT data.
The analysis of these documents shows the neces-
sity to clarify and define the extent of PNT data,
which should be delivered by a maritime PNT Unit.
In a preliminary design, the following parameters
should be considered:
1 Position: It mainly contains the longitude and
latitude for maritime navigations. Because ves-
sels can usually be found close to the sea level,
the height information is usually not provided
as standard output parameter.
2 Under keel clearance (UKC): Instead of the
height information, the UKC is the relevant
maritime output parameter. It is defined as the
distance between the lowest point of the ship
(e.g. the keel) and the ground of the sea.
3 Velocity: The magnitude and direction of a ve-
locity vector can be described by Speed over
Ground (SOG) and Course over Ground
(COG). Because of their physical principles,
speed sensors like e.g. electro-magnetic logs
can only measure the speed through water
(STW), and therefore STW is also a parameter
which a PNT Unit could deliver.
4 Attitude: Generally the orientation of the ship
in the horizontal plane is reported. Here one
needs to distinguish between the orientation
with respect to the true north (true heading) and
with respect to the magnetic north (heading).
For future applications, the other attitude an-
gles, namely roll and pitch, could also be re-
quired.
5 Timing: UTC time needs to be delivered.
After the clarification of the PNT output parame-
ters, further additional requirements on the PNT Unit
will be discussed in the following.
In [1], the robustness of all e-navigation systems
is requested. In order to fulfill this requirement, a
definition of robustness needs to be given. We inter-
pret robustness as the ability of a system to provide
the output data according to their specification under
changing application conditions and in cases of ex-
ternal disturbances (interferences, jamming, atmos-
pheric influences). The robustness shall therefore be
applicable to the realization of the basic functional-
ity (output data with required accuracy) and integrity
functionality.
In [1] it is furthermore stated, that requirements
for redundancy, particularly in relation to position
fixing systems, should be considered. Redundancy in
a general meaning can be seen as the provision of an
alternative system to support critical system func-
tionalities. Within the Recommendation R-129 on
GNSS Vulnerability and Mitigation Measures [10]
IALA has given a classification of alternative navi-
gation systems in relation to their aims:
1 A redundant system provides the same function-
ality as the primary system, allowing a seamless
transition with no change in procedures.
2 A backup system ensures continuation of the
navigation application, but not necessarily with
the full functionality of the primary system and
may necessitate some change in procedures by
the user.
151
3 A contingency system allows safe completion of
a manoeuvre, but may not be adequate for long-
term use.
For the introduction of additional sensors to the
integrated PNT Unit, this classification scheme
needs to be considered.
2.2 Standard sensors for the provision of PNT data
According to the carriage requirement demanded by
the IMO Safety of Life at Sea (SOLAS) convention
[11], following sensors (see Table 1) should be used
in maritime applications. An overview of the sensors
and a discussion of the related standards can be
found in [12]. Therefore here only a table with the
list of standard sensors with typical output data and
realizations are given.
TABLE 1. STANDARD PNT SENSORS
Sensor
Output
Typical realization
Speed log [13]
Speed through water
(STW)
Electromagnetic
logs
Speed over ground
(SOG)
Doppler logs
Compass [14]
[15]
True heading
Gyrocompass
Magnetic heading
Magnetic compass
Electronic Posi-
tion Fixing Sy
s-
tem (EPFS)
Position
Speed over Ground
(
SOG)
Course over ground
(COG)
Time
GPS/GLONASS
DGPS/DGLONASS
receiver and antenna
Transmitting
Heading
Device
[16] (THD)
Heading
GNSS multi- anten-
na system
Echo sounder
[17]
UKC
Sonar
Rate of Turn
Indicator [18]
(ROTI)
ROT
Mechanical Gyro
3 ACCURACY ASSESSMENT AND
INTEGRITY MONITORING
3.1 Integrity Definition
One of the key tasks of an integrated PNT Unit is the
additional provision of integrity messages to the us-
er. Within the specification of requirements for fu-
ture GNSS [8] IMO has given the following defini-
tion of in integrity: The ability to provide users with
warnings within a specified time when the system
should not be used for navigation. Within that defini-
tion the alert limit specifies the applicable threshold
differing between an unusable or unusable system.
The time to alarm (TTA) describes the acceptable
time duration between occurrence of an intolerable
error and the provision of the related integrity mes-
sage. The remaining integrity risk (IR) specifies the
tolerable probability, that the assessment of accuracy
by integrity monitoring is erroneous. This definition
implies an integrity monitoring process wherein the
accuracy of the given navigation parameter needs to
be estimated in real time and compared against a
given threshold (alert limit). In other words, an er-
ror estimation of all navigation parameters needs to
be performed onboard a vessel.
Integrity monitoring as it is defined for Integrated
Navigation Systems (INS) consists of the following
three sequential steps [19]:
− Plausibility check
The plausibility check tests whether the sensor
raw data or derived navigational result falls into
predefined value range. The plausibility check is
normally carried out at receiver side in order to
test the usability of the sensor data.
− Validity check
The validity is tested by comparing the sensor da-
ta or derived navigational results with formal and
logical criteria as well as by checking the correct-
ness of the data format. The validity check is
normally carried out at the sensor site in order to
ensure the proper operation of the sensor.
− Compatibility check
Once a specific parameter can be provided by
more than one sensor, different sensor data can be
compared to test the compatibility. A significant
discrepancy between different sensor data implies
the failure of at least one of these sensors. The
upper bound for deviation should be defined ei-
ther a priori or in real-time according to the pre-
vious measurements. The compatibility test
should be carried out before sending the sensor
data to integration algorithm.
The primary aim of plausibility, validity and
compatibility checks is the detection of errors. The
assessment of accuracy requires the implementation
of suitable integrity monitoring functions.
3.2 Existing Integrity Monitoring approaches
In this section existing integrity monitoring ap-
proaches defined for maritime navigation will be de-
scribed and the seen gaps will be discussed.
3.2.1 Integrity monitoring for GNSS
3.2.1.1 Receiver Autonomous Integrity Monitoring
(RAIM)
RAIM is a technique, which can be applied in
GNSS receivers to assess the integrity of navigation
signals [8], if more than 5 GNSS signals are tracked
[8]. RAIM has two-fold tasks, first is to check the
occurrence of a failure, and second is to identify er-
roneous satellites. RAIM can be done by using only
the incoming measurement of current epoch or em-
ploying also measurements of previous epochs. The
152
former approach needs the redundant measurement,
and the performance is dependent on the number of
satellites in view. In [20], several classic “snapshot”
algorithms are introduced.
For a maritime GNSS receiver a RAIM is re-
quested [21], however, neither the algorithm is spec-
ified nor it is defined how the user should react on
the three possible outputs: safe, caution, and unsafe.
Referring to a “snapshot” RAIM, the a priori
knowledge of the observation errors is needed by all
existing algorithms. A proper determination of this
term for maritime navigation is an open task.
3.2.1.2 IALA Beacon DGNSS
GNSS integrity-monitoring services are usually
part of augmentation services which also provide
DGNSS corrections. The reason for this shared ac-
tivity is the similarity of the infrastructure required
for DGNSS and integrity monitoring.
The IALA Beacon Differential GNSS service is a
standardized technique for maritime use [22]. The
implemented integrity monitoring assesses primarily
the integrity of the service itself, but can also include
parts of GNSS assessment. At reference station dif-
ferent thresholds are applied to the observed HDOP
or determined range and range rate corrections to es-
timate the usability of service or single GNSS sig-
nals. Additionally the IALA Beacon DGNSS is
equipped with one or more integrity monitoring sta-
tions operating as virtual user at known location.
Only in case that the observed DGNSS position er-
ror is below the allowed error threshold, the IALA
Beacon DGNSS distributes a flag bit to indicate the
usability of the service. The health status of a satel-
lite is provided indirectly by embedding the “do not
use” flag in the transmitted PRC and RRC for an
unhealthy satellite.
Spatial decorrelation of provided range and range
rate corrections, differences in satellite visibility be-
tween station and user sites, and user specific recep-
tion conditions (multipath, interferences) are the
main causes for discrepancies between the estimated
and real DGNSS positioning performance.
3.2.1.3 Maritime GBAS (RTK)
The preliminary integrity monitoring of the phase
based Maritime GBAS [23] follows the IALA Bea-
con DGNSS concept. Hence an integrity monitoring
station (IM) is installed in the service area (ca. 10-
20km) of the reference station (RS). At both stations
integrity monitoring procedures are processed in
three steps. In the first step the RS and the IM evalu-
ate the quality of the received GNSS signals on the
basis of quality parameters like phase and code
noise. In a second step a GNSS RAIM based posi-
tion determination is realized. The results of both
steps are compared with pre-specified thresholds to
assess the usability of single GNSS signals as well
as the usability of the service of RS and IM station.
Afterwards the signals without abnormalities from
the RS and IM are used to calculate the IM position
by carrier phase based differential techniques (RTK).
The position error is derived by comparison of the
computed with the exactly known position of the
IM. If the accuracy requirements of IMO for port
vessel operation are fulfilled, the service can be con-
sidered as usable.
Finally the M-GBAS logically combines all gath-
ered integrity information at RS and IM site gained
in the previous steps to generate the RTCM 3 mes-
sages for the provision of augmentation data and re-
lated integrity data (message 4083).
3.2.1.4 Integrity monitoring for INS
An INS offers integrated and augmented func-
tions to support system tasks like collision avoid-
ance, route planning and route monitoring. Currently
an INS is not a mandatory system, but if an INS is
installed onboard a vessel it is accepted monitoring
is considered as an intrinsic function of the INS.
The currently valid INS standard is based on IMO
resolution MSC.86 (70) [24] and is specified within
the IEC-61924 standard [19]. A task oriented con-
cept is already introduced in a new resolution
MSC.252 (83) [25], but the specification within the
related IEC standard is not yet published. Therefore
our analysis is based on IEC-61924 standard on-
ly. In IEC-61924 standard, plausibility, validity and
compatibility check approaches are introduced. IEC-
61924 standard suggests the compatibility check for
the following navigational parameters.
1 Position: comparison with a second EPFS; using
RAIM GNSS function; Dead Reckoning (DR) us-
ing the ship’s heading and speed measuring de-
vice
2 Heading: comparison with a second heading sen-
sor and a course over ground sensor
3 SOG: comparison with a second SOG sensor,
with speed through water sensor and with SOG
from the EPFS (GNSS)
4 Time: comparison with a second time sensor and
with the internal INS clock
5 UKC: comparison with a second depth sensor and
with data derived from ships position and elec-
tronic navigational charts (ENC)
3.3 Demand on specification and development
The Position, as the most important PNT infor-
mation, currently is measured by only two separate
receiver/antenna GNSS devices. Integrity monitor-
ing is restricted to a comparison of the positions
determined by these two receivers. Due to the same
measurement technique, system and propagation er-
153
rors of both GNSS devices underlie the similar error
in the measurement and position domain.
An estimation of the actual position error is cur-
rently not performed onboard a vessel. Indirectly it is
assumed, that the horizontal position error is < 100
m, when using GNSS standard positioning service,
and <10 m, when using IALA Beacon DGNSS ser-
vice.
Although positioning accuracy requirement on
GNSS are specified with respect to different op-
erational areas [26][8], the integrity monitoring is
currently performed by using fixed or user se-
lectable thresholds. In order to use these areas for fu-
ture integrity monitoring applications, these areas
and the intersection from one area to another need to
be clearly specified in an appropriate way (e.g. in
ENC charts).
Furthermore integrity specifications for the other
PNT parameter (e.g. SOG, COG, time) besides posi-
tion have to be specified. The necessity of an opera-
tional area dependent accuracy and integrity specifi-
cation for these parameters needs to be discussed
Analyzing the existing integrity monitoring ap-
proaches with respect to the identified user needs [1]
a demand on the development of an enhanced integ-
rity monitoring for all relevant PNT data within an
integrated PNT Unit can be deduced. Such a PNT
Unit should use sensor and data fusion methods to
ensure the provision of PNT output data with the de-
sired accuracy and to ensure an overall integrity
monitoring for these output data. For both func-
tionalities a higher degree of standardization is de-
sired in order to achieve comparable results for their
harmonized application.
4 PROPOSED ARCHITECTURE OF
INTEGRITY MONITORING IN PNT UNIT
For a PNT Unit, integrity monitoring can be carried
out in three sequential steps. The first step is an in-
dividual sensor data test. The second step is the
compatibility test of similar data from different sen-
sors. The third step is the fault detection and identi-
fication in the integration algorithm. A general in-
tegrity monitoring approach is depicted in Figure 2.
4.1 Integrity monitoring for GNSS
Actually within an INS, position integrity monitor-
ing is performed by comparing the calculated posi-
tion with a second EPFS, by using RAIM GNSS
function and by using DR technique. Possible im-
provements in a PNT Unit will be elaborated as fol-
lows.
Figure 2. General integrity monitoring approaches in the integrated
sensor system
4.2 Compatibility check for redundant GNSS
systems
GNSS redundancy can be achieved using a second-
ary GNSS device, using multiple civilian frequencies
and using multiple GNSS constellations.
Once the major GNSS device (antenna or receiv-
er) is out of use, the second GNSS device can fully
take the function of the major GNSS device. How-
ever, the redundant GNSS device is also affected by
the errors related to the radio signal. In this sense,
the significance of a redundant GNSS device is re-
flected during the internal failure of the major GNSS
device.
Modern GNSS satellites will facilitate more than
one signal. Civilian code data will also be encoded
into the carrier signal besides L1 signal at future
GNSS satellites. The additional civilian code data
will offer the same functionalities like the SPS ser-
vice. Due to different carrier signals, the other carri-
er signals might not suffer from the same interfer-
ence or jamming or propagation effects as L1 signal.
Also, the channel failure (loss of lock or cycle-slips,
etc.) for L1 signal might not occur simultaneously on
the other frequencies. Nevertheless, errors due to
space atmosphere and signal propagation will influ-
ence all the carrier signals of a satellite. Hardware
failure of receiver or antenna might also challenge
the reception of all carrier signals.
Two or more full-operational GNSS constella-
tions could serve as redundancy for each other, as
they realize same functions in maritime navigation
as specified in [26]. However, GLONASS and GAL-
ILEO are not yet fully operational. Future GNSS re-
ceivers and corresponding antennas allow the recep-
tion and processing of multiple GNSS signals,
however, a hardware failure can cause the loss of all
GNSS signals.
154
4.3 Compatibility check with backup systems
Systems like e-Loran (enhanced Long Range Navi-
gation) or R-mode (Ranging mode) facilitate the
functions for positioning. The e-Loran system can
also be used for time determination, so that these
systems could serve as backup for GNSS [27].
Compared to GNSS, e-Loran signals are transmitted
at lower frequency with higher power and hence it is
not easy to be jammed especially not by the same
GNSS jammers. It relies on the radio signal propa-
gated over ground and hence does not suffer from
the same errors in the propagation path from sky like
GNSS. So the future of e-Loran as a terrestrial back-
up for GNSS with a large coverage area is currently
an open question. Also, the fulfillment of the future
maritime requirements with respect to the accuracy
is an issue.
In [2], the R-Mode is seen as a possible novel
variant of positioning technique using terrestrial
signals. The idea is to use existing communication
channels and append their functionality by sending
an additional timing signal. From the time differ-
ence between signal transmission and reception, the
ship should be able to determine its position. The
advantage of this idea would be that at least partially
existing infrastructure could be used. Currently this
is still only an idea, where the proof of concept needs
to be shown.
4.4 Compatibility check with contingency system
The DR is a frequently-used technique to predict the
position using SOG and COG information. In mari-
time navigation, COG information is usually approx-
imated by compass. DR is independent of the radio
signals and hence still works during GNSS outage.
In a sensor fusion system, DR is already an implicit
function and does not need to be separately imple-
mented.
Another contingency system can be constructed
by introducing the inertial sensor. Also the inertial
sensors are still not standard sensors in maritime
navigation, they are drawing more and more interests
due to the independency of radio signals, the short-
term high accuracy, the high output rate and the de-
creasing price. An Inertial Measurement Unit
(IMU), which is composed of three orthogonal ac-
celerometers and gyroscopes, can offer the naviga-
tion parameters like position, velocity, rate of turn
and attitude. For this reason, introducing an IMU
allows the integrity monitoring for relevant sensors.
A drawback is that the IMU cannot work alone for
long-term use and hence needs to be integrated with
other sensors.
4.5 RAIM
Classic RAIM algorithms can also be applied to the
maritime navigation. The only problem is the deter-
mination of the a priori measurement error for pseu-
doranges. This relies on standardized specification
for different operation areas. If this is not available,
empirical values have to be used. As an added value
of the integrated system, an enhanced RAIM aided
by the antenna dynamics can be implemented. The
classic “snapshot” RAIM is based on the received
pseudorange data of current epoch. Once the antenna
dynamics can be determined, the antenna position
can be predicted from the position of last epoch.
This makes the pseudoranges of current epoch pre-
dictable. The predicted pseudoranges serve as addi-
tional observation to enhance the integrity monitor-
ing.
4.6 Integrity monitoring for other independent
navigation sensors
Except for GNSS, the error estimation for navigation
sensors is difficult as the sensor raw data is not di-
rectly processed. However, the sensor fusion algo-
rithm makes it possible. Once the error estimation of
one sensor data is available, the fusion algorithm al-
lows the error estimation for other relevant sensors.
For example, if the error of GNSS- based SOG
can be properly assessed, the error analysis of speed
log is also possible.
4.7 Integrity monitoring in integration algorithm
The plausibility tests, validity tests and compatibility
tests are suitable for detecting gross sensor failure
but not sensitive for slight error, time-variant errors
and drifts. The Kalman filter-based algorithm could
offer high sensitivity of detecting these errors. Integ-
rity monitoring based on Kalman filter can be cate-
gorized into the following approaches [27].
4.7.1 Kalman filter estimates (bias check)
In a Kalman filter, the errors of navigation pa-
rameters can be estimated. If an estimated error is
significantly larger than the error level specified by
the manufacturer, it is likely to be a failure in the
sensor.
4.7.2 Innovation-based approaches
The innovations indicate the consistency of the
actual measurements and the measurements predi-
cated by state estimates. Innovation filtering may be
used to detect large discrepancies immediately,
whereas innovation sequence monitoring enables
smaller discrepancies to be detected over time.
155
4.7.3 Residual-based approaches
The above-mentioned innovation filtering and se-
quence monitoring can also be expanded to residu-
als. Residuals have a smaller covariance than inno-
vation, making them more sensitive for error
detection [27]. The only shortcoming is that the pro-
cessing of residuals is not an essential part of a Kal-
man filter routine and needs extra computing time.
4.7.4 Parallel solution of multiple sub-filters
Parallel-solutions integrity monitoring maintains
a number of parallel navigation solutions or sub-
filters, each excluding data from one sensor or radio
navigation signal. Each additional navigation solu-
tion is compared with the main filter using a con-
sistency test. A significant inconsistency indicates a
fault in the sensor or signal omitted from main filter.
The system output is then switched to the solution
omitting the faulty sensor or signal. The main draw-
back lies in the increased computational burden and
hence this technique is preferably used for failure
identification rather than failure detection
4.8 Initial design of integrity monitoring in PNT
Unit
The major integration strategy of a PNT Unit is the
integration of GNSS and IMU. An initial design of
sensor integration in a PNT Unit is depicted in Fig-
ure 3. The plausibility check and compatibility
check are marked at the corresponding positions in
the data flow.
Figure 3. Integrity monitoring in the architecture of a PNT unit
The plausibility check is carried out for gyrocom-
pass, speed log and IMU output. Taking the dynamic
properties of a specific ship into account, the
threshold value can be derived for plausibility
check.
The compatibility check is first applied to GNSS
in order to determine the major GNSS antenna (also
the constellation and signal, if not only one constel-
lation or one signal is to be used). The data of major
GNSS antenna will be integrated with other sensors.
Both GPS and IMU offer velocity parameters and
hence can be compared with speed log. IMU also
outputs attitude (heading) and hence can be com-
pared with gyrocompass.
The Kalman filter enables the bias check, innova-
tion check and residual check. The prerequisite is
a stable operation of the filter mechanism. As a
stochastic system, the filter performance is based on
the modeling of the observation and dynamic model,
the reasonable a priori knowledge of the observa-
tion and dynamic errors, and most important, the real
application scenario. It is not a trivial task to adapt
the filter to all potential unexpected situations, and
hence it is necessary to test the stability of the filter.
The stability test can be done either using the in-
ternal parameters of the filter, or by checking the
compatibility of filter results with the results from
other stand-alone sensors.
4.9 Integrity output from a PNT Unit
According to the previous analysis, the following
integrity parameters will be supported by a PNT
Unit.
TABLE 2. OUTPUT FROM A PNT UNIT
___________________________________________________
PNT Plau- Vali- Compatibility Estimated
Output sibility dity Error
___________________________________________________
Position x x Redundant GNSS x
Other radio-navigation
system
SOG x x GNSS Doppler, Log, x
IMU,
COG x x GNSS, IMU
Attitude x x Compass x
multi-antenna system
Rate of x x IMU x
Turn heading variation
with time
Time x x Other time sensors
internal clock of an INS
___________________________________________________
5 SUMMARY
This paper focuses on a maritime integrated PNT
Unit as the on-board part of maritime PNT system.
The aim of the PNT Unit is the robust provision of
position, navigation and timing information in ac-
cordance with the performance requirements of the
actual operational region. IALA has introduced the
term PNT device [2] as “a device using any availa-
ble IMO recognized radio navigation systems simul-
taneously to provide the best electronic position fix
for the ship”. The PNT Unit, proposed in this paper,
follows a more general approach in combining
available radio navigation systems and their augmen-
tation services with other on-board PNT sensors.
The core of the PNT Unit is a processing system,
156
which combines all available PNT sensors. The PNT
Unit is on the one hand part of the integrated PNT
System and on the other hand part of the on-board
INS.
Focusing on integrity for PNT information we
have analyzed the state-of-the-art integrity monitor-
ing approaches with respect to the identified user
needs. Based on this, a demand on the development
of an enhanced integrity monitoring for all relevant
PNT data within an integrated PNT Unit can be de-
duced. Such a PNT Unit should use sensor and data
fusion methods to provide PNT output data and im-
prove integrity monitoring for these output data. For
both functionalities a high degree of standardization
is desired in order to achieve comparable and relia-
ble results for their harmonized application.
Subsequently we have introduced a preliminary
integrity monitoring concept for a PNT Unit which
also includes additional sensors in order to deliver
redundancy, backup or contingency functionality.
Finally it should be stated that this paper can on-
ly be seen as a starting point towards the realization
of an integrated onboard PNT Unit for maritime ap-
plications. In a next step, after consolidation of the
architecture, we plan to develop a demonstrator sys-
tem of a PNT Unit.
REFERENCES
[1] IMO, NAV 54/25 Annex 12 Draft strategy for the devel-
opment and implementation of e-Navigation, 2008.
[2] IALA, World Wide Radio Navigation Plan, Edition 1,
2009.
[3] IMO, Resolution MSC.112(73): Adoption of the Revised
Performance Standards for Shipborne GPS Receiver
Equipment, 2000.
[4] IMO, Resolution MSC.113(73): Adoption of the Revised
Performance Standards for Shipborne GLONASS Receiver
Equipment, 2000.
[5] IMO, Resolution MSC.114(73): Adoption of the Revised
Performance Standards for Shipborne DGPS and
DGLONASS Maritime Radio Beacon Receiver Equipment,
2000.
[6] IMO, Resolution MSC.115(73): Adoption of the Revised
Performance Standards for Shipborne GPS/GLONASS
Combined Receiver Equipment, 2000.
[7] IMO, Resolution MSC.233(82): Adoption of the Perfor-
mance Standards for Shipborne Galileo Receiver Equip-
ment, 2006.
[8] IMO, Resolution A.915(22): Revised Maritime Policy and
Requirements for A Future Global Navigation Satellite Sys-
tem (GNSS), 2001.
[9] IMO, NAV 56/WP.5 Development of an e-Navigation
Strategy Implementation Plan, 2010.
[10] IALA, Recommendation R-129: On GNSS Vulnerability
and Mitigation Measures Edition 2, 2008.
[11] IMO, SOLAS Chapter V: Safety of Navigation, 2002.
[12] R. Ziebold, Z. Dai, T. Noack, und E. Engler, “Concept for
an Integrated PNT-Unit for Maritime Applications,” Pro-
ceedings of the 5th ESA Workshop on Satellite Navigation
Technologies. Navitec 2010, 8.-10. Dec. 2010, Noordwijk,
The Netherlands.
[13] IMO, Resolution MSC.96(72): Adoption of Amendments
to Performance Standards for Devices to Measure and Indi-
cate Speed and Distance, 2000.
[14] IMO, Resolution A.424(XI): Performance Standards for
Gyro Compasses, 1979.
[15] IMO, Resolution A.382(X): Magnetic Compasses Car-
riage and Performance Standards, 1977.
[16] IMO, Resolution MSC.116(73): Performance Standards
for Marine Transmitting Heading Devices (THDs), 2000.
[17] IMO, Resolution A.224(VII): Performance Standards for
Echo- Sounding Equipment.
[18] IMO, Resolution A.526(13): Performance Standards for
Rate-Of-Turn Indicators, 1983.
[19] IEC, IEC 61924 Maritime navigation and radiocommuni-
cation equipment and systems – Integrated navigation sys-
tems – Operational and performance requirements, methods
of testing and required test results, 2006.
[20] B.W. Parkinson und J.J. Spiker, Global Positioning Sys-
tem: Theory and Applications: Volume II, American Insti-
tute of Aeronautics & Astronautics, 1996.
[21] IEC, Maritime Navigation and Radiocommunication
Equipment and Systems – Global Navigation Satellite Sys-
tems (GNSS) – Part 1: Global Positioning System (GPS)
Receiver Equipment, Performance Standards, Methods of
Testing and Required Test Results.
[22] IALA, Recommendation R-121: The Performance and
Monitoring Of DGNSS Services in the Frequency Band
283.5 – 325 KHz, 2004.
[23] D. Minkwitz und S. Schlueter, “Integrity Assessment of a
Maritime Carrier Phase Based GNSS Augmentation Sys-
tem,” in Proceedings of ION GNSS, Portland, USA: 2010.
[24] IMO, MSC.86(70): Adoption of new and amended per-
formance standards for navigational equipment, 1998.
[25] IMO, Resolution MSC.252(83): Adoption of the Revised
Performance Standards for Integrated Navigation Systems
(INS), 2007.
[26] IMO, Resolution A.953(23): World-Wide Radio Naviga-
tion System.
[27] P.D. Groves, Principles of GNSS, Inertial, and Multi-
Sensor Integrated Navigation Systems (GNSS Technology
and Applications), Artech House Publishers, 2007.
