International Journal

on Marine Navigation

and Safety of Sea Transportation

Volume 3

Number 1

March 2009

Manoeuvring Simulation on the Bridge for

Predicting Motion of Real Ships and as

Training Tool in Ship Handling Simulators

K. Benedict, M. Kirchhoff, M. Gluch & S. Fischer

Hochschule Wismar University of Technology, Warnemunde, Germany

M. Baldauf

World Maritime University, Malmo, Sweden

1 INTRODUCTION

International sea transport has growing rapidly dur-

ing the period of the last decade. Ships became larg-

er and wider and its container capacity is still in-

creasing to 12.000 TEU and even more. To navigate

such vessels safely from port to port and specifically

within the ports from and to the dedicated berths

more and more enhanced computer-based systems

are installed on the ships navigational bridge to sup-

port the pilot, the master and his navigating officers

as well.

Investigations are ongoing to integrate features

for new manoeuvring and steering equipment such

as azimuth propellers or waterjets and in parallel to

enhance the predictions of the complex own ship

motions taking into account the use of the controls in

time.

Prediction tools are very helpful and are already

in use on ships for a long time, beginning with trial

modes in ARPA radars up to curved headline over-

lay in ECDIS. However, the simplification of these

predictions allows restricted use only based either on

estimated future courses & tracks or on the simple

integration of the current ship motion not including

the immediate response on changes of rudder and

engine.

New concepts for on board displays and simula-

tion tools were developed in research projects fund-

ed by the German Federal Ministry of Education and

Research together with partners from manufacturers

like SAM Electronics Hamburg and INTER-

SCHALT/AVECS.

A prediction tool was developed to simulate the

ships motion with complex dynamic models in fast

time and to display the ships track immediately for

the intended or actual rudder or engine manoeuvre.

ABSTRACT: International sea transport has growing rapidly during the period of the last decade. Ships be-

came larger and wider and its container capacity is still increasing to 12.000 TEU and even more. To navigate

such vessels safely from port to port and specifically within the ports more and more enhanced computer-

based systems are installed on the ships navigational bridges. Prediction tools are very helpful and already in

use on ships for a long time. However, the simplification of existing predictions allows restricted use only and

do not include the immediate response on changes of rudder and engine. Within this paper investigations into

the feasibility and user acceptance of newly developed layout of navigation display will be introduced and se-

lected results of simulation studies testing the influence on manoeuvre performance dependent on different

kind of prediction functions will be discussed. Examples will be given for results from test trials in the full

mission ship handling simulator of the Maritime Simulation Centre Warnemunde and a concept for the appli-

cation of the developed .tools for purposes of collision avoidance is described.

These simulations are based on input from the ships

actual sensors via the Voyage Data Recorder and

furthermore from diagnosis tools analysing the sta-

tus of the manoeuvring facilities and providing in-

formation in case of failures, e.g. reduced engine

power or larger rudder response time due to mal-

functions of the equipment.

This tool can be used both for real ships operation

on board but also for the effective training in simula-

tors because of its unique advantage that the conse-

quences of manoeuvring commands can be seen

immediately before the ship has even changed her

motion.

Within this paper investigations into the feasibil-

ity and user acceptance of the new layout of naviga-

tion display will be introduced and selected results

of simulation studies testing the influence on ma-

noeuvre performance dependent on different kind of

prediction functions will be discussed. Examples

will be given for results from test trials in the full

mission ship handling simulator of the Maritime

Simulation Centre Warnemunde.

2 STATE OF THE ART AND NEW APPROACH

The role of computer based simulation is increasing

on the ships bridge, especially for manoeuvre plan-

ning and for collision avoidance. Prediction tools are

very helpful and already in use on ships for a long

time. Well known is the so called Trial Manoeuvre

mode in ARPA radars to be used in order to analyse

future encounter situation for selected relevant

course and speed alternatives to deck potential colli-

sion avoidance strategies.

With the emerging Electronics Chart and Infor-

mation Systems ECDIS new tools were introduced

for supporting voyage planning by means of

manoeuvring characteristics. For controlling the ship

on her route the future track of the ship was shown

as a so called “curved headline” overlay in ECDIS.

However, theses prediction are very simple only

based either on new constant course and speed val-

ues as in the ARPA trial function or on estimated fu-

ture courses & tracks based on the simple integration

of the current ship motion parameters as rate of turn

and speed components to be considered as constant.

The simplification of these predictions allows re-

stricted use only. That is why new concepts for on

board displays and simulation tools were developed

using an innovative approach which includes the

immediate response on changes of rudder and engine

commands for the display of the future track.

This approach was investigated in research pro-

jects, dedicated on the one hand to the further devel-

opment of user interfaces on ships navigational

bridges and to investigations into potential im-

provements for manoeuvring assistance on the other

hand.

A prediction tool was developed to simulate the

ships motion with complex dynamic models in fast

time and to display the ships track immediately for

the intended or actual rudder or engine manoeuvre

(Benedict, Baldauf et al 2007). Generally there are

two areas of application of such a prediction tool. It

can be seen both as training tool for ship manoeu-

vres and to be used as assistance tool on board ves-

sels:

− Training Tool: The prediction of ships motion as

an immediate response could be an excellent

method to demonstrate the results of changes or

alternatives of using manoeuvring control devices

as for instance propellers, rudders or thrusters.

This is of increasing importance specifically for

the growing complexity of manoeuvring control

systems starting from simple one-propeller and

middle rudder, via twin propellers with double

rudder up to new azimuth propellers which can be

turned by 360° (there are ships with even four of

these sophisticated thrusters).

− Assistance Tool: Predictions as elements of on

board displays can be used as in the loop control

elements to steer the ship manually but supported

by the future track or speed indication in the EC-

DIS interface.

One crucial problem for the prediction is the ac-

curacy of the simulation. In the mentioned projects a

very sophisticated approach was used to represent

the ships’ dynamic by very extensive equations very

similar to those used in Full Mission ship handling

simulators. The parameters of the equation of mo-

tion will be estimated by an extra fast time simula-

tion program and a data analyser already used for

tuning of the hydrodynamic models in the ship han-

dling simulator. These methods will be described in

the following chapters and examples will be given

for results from test trials in the full mission ship

handling simulator of the Maritime Simulation Cen-

tre Warnemuende upgraded in 2007/2008.

This Simulation Centre accommodates six simu-

lators embracing a common network and comprised

of four ship-handling bridge systems with differing

levels of equipment, a ship's engine system and a

VTS simulation facility.

The interaction of many of the single simulators

is one of the unique features of the MSCW: they can

be interfaced to form a big scenario comprising all

simulators and connecting e.g. the big bridge 1 with

the full mission engine simulator. (Benedict 2000).

Figure 1: Maritime Simulation Centre at Warnemuende

(MSCW) which comprises three interfaced simulator segments

for ship handling, ship engine and VTS

3 APPROACH FOR PREDICTION TOOL

3.1 Ship dynamic model and Technological Setup

The following equation of motion was used as math

model for the ships dynamic:

( )

rxrvumX

−−=



( )

rxruvmY



++=

( )

ruvmxrIN

++=



On the right side are the effects of inertia where u

and v represent the speed components in longitudi-

nal and transverse direction x and y, r is the rate of

turn of the ship. The ships mass is m and x

is the

distance of centre of gravity from the origin of the

co-ordinate system, I

is the moment of inertia

around the z-axis.

The ships hull forces X and Y as well as the yaw-

ing moment N around the z-axis are on the left side.

Their dimensionless coefficients are normally repre-

sented by polynomials based on dimensionless pa-

rameters, for instance in the equation for transverse

force Y and yaw moment N given as the sum of

terms with linear components Nr, Nv, Yr and Yv

and additional non-linear terms. Other forces as for

instance rudder forces and wind forces are expressed

as look up tables. There are additional equations for

the engine model, where are also look up tables to

represent automation systems characteristics. The

solution of this set of differential equations is calcu-

lated every second; some internal calculations are

even done with higher frequency.

The Input output relations are shown in Figure 2.

The inputs consist of controls, the states and the data

for the environmental conditions in the three blocks

on the left side. The core module Simula-

tion/Prediction is in the centre of the figure. Addi-

tionally there is an input of the Ships condition pa-

rameters. They are normally fixed but in case of

malfunctions they might change, e.g. reducing the

rudder turning rate or maximum angle. The results

from the Simulation block are transferred to be dis-

played in ECDIS or Radar.

Figure 2. In-/Output concept for prediction process and data

flow

In Figure 3 the more technological setup of the

structure of modules is described. A commercial

IMO-proven Voyage Data Recorder (VDR) plays

the role of data collector for the controls, states and

environmental parameters measured by the ship sen-

sors.

Figure 3. Modules & data sources and sinks

After pre-processing the data will be stored in

Shared Memory 1, together with the condition pa-

rameters which will be provided by a diagnosis sys-

tem. This system continuously checks the ships and

engine conditions. From this memory the data are

available for other modules:

− The Simulation Prediction Module uses the data

from Shared Memory 1 to predict the ships track

and speed for a certain time period. The results

are sent to Shared Memory 2.

− The Presentation Module uses the data both to

display the actual position and from Shared

Memory 2 to display the future track.

− The Prediction parameters are controlled by an

user interface integrated in the Presentation mod-

ule with regard to predicting cycle and length of

track.

3.2 Presentation of dynamic Predictions in ECDIS

environment

For a compact presentation of information to the

captain, pilot and responsible navigating officer re-

spectively a new layout of a conning display was de-

signed and implemented into the equipment installed

on an integrated navigation system. The display lay-

out contains an overlay of ECDIS and CONNING

information together with the prediction (figure 4).

In the centre the ECDIS information in Head up

Mode together with motion parameter for longitudi-

nal speed (10.1.kn and transverse speed (0.1 kn) as

well as a circle segment with the rate of turn to STB

((4.0 °/min) is shown. The ships position is dis-

played in the centre of the ECDIS as ships contour

where the track prediction can be indicated as

curved track or as chain of contours for the selected

prediction time. The prediction parameters as range

or interval of presentation can be set in the control

window at the right side.

Figure 4. Layout concept for Manoeuvring Prediction in EC-

DIS

Figure 5. Comparison of methods based on different tracks

In the centre the ECDIS information in Head up

Mode together with motion parameter for longitudi-

nal speed (10.1.kn and transverse speed (0.1 kn) as

well as a circle segment with the rate of turn to STB

((4.0 °/min) is shown. The ships position is dis-

played in the centre of the ECDIS as ships contour

where the track prediction can be indicated as

curved track or as chain of contours for the selected

prediction time. The prediction parameters as range

or interval of presentation can be set in the control

window at the right side.

The predicted track for the simplified prediction

is shown as red curve (here shown in black to star-

board): According to the actual rate of turn to star-

board the conventionally predicted track is presented

as a circle segment to the right side as track for the

time range of 5 min with a speed of 10.1 kn.

The dynamic prediction with the full simulation

model is shown as blue curve (here shown in black

to port). This dynamic prediction reflects the setting

of rudder and propeller parameters shown in the left

bottom window: The two rudders of the ferry used in

this example are set to 14° Port and the Engine Or-

der Telegraph for the two controllable pitch propel-

lers are set to 100% representing 159.8 rpm of the

propeller. The actual pitch status is 53 and 54 re-

spectively. This interface allows for a presentation

of dynamic predictions of steering and stopping

characteristics as an immediate response according

to the current steering handle or engine order tele-

graph position.

3.3 Investigations into the effects of predictions on

ship handling

For the purpose of testing the potential effects of

such enhanced prediction tool it was implemented in

the INS equipment of the large full mission simula-

tor bridge of the Ship handling simulator of MSCW.

For trials to test the effects of such a tool on the

navigators behaviour a sample of a PanMax contain-

er vessel with Loa =294m, Boa=32,2m, Draft=12m

and a Displacement of 74.000 t is used in the simula-

tion experiment. The container vessel is equipped

with one fixed propeller and one balanced rudder

blade. Additionally one bow thrusters is available

for manoeuvring the ship.

Figure 6: sample of a result from experimental trial to investi-

gate effects of predictions tool

A first basic test scenario was developed and im-

plemented. The task is to steer a vessel from a berth

into a fairway to leave a harbour area. The scenario

is used for trials with participants who are not famil-

iar with the selected ship. The task to be performed

is to safely manoeuvre the ship into the fairway.

Each participant started with a trial to become famil-

iar with the ships behaviour and without using the

prediction tool. The second run of the test contains

the same task but were performed in another area of

the port. An example of a trial using the prediction

tool is given in the following figure.

After the first series of simulation runs there are

already some tendencies regarding the effects that

can clearly be identified.

− There are successful manoeuvres, when the pre-

dictor was used; even if the test person has no ex-

periences for the specific ship and the area.

− The execution time for the manoeuvre is smaller

when the predictor is used.

− The number of orders (for engine, thrusters and

rudder) decreases with the use of the predictor.

To proof these tendencies the trials will be con-

tinued with cadets during the next semesters. Then

more detailed investigations with more detailed sta-

tistical analysis will be performed.

4 APPLICATION OF THE PREDICTION

MODULES FOR ON BOARD COLLISION

AVOIDANCE

The prediction algorithm and the technical setup is

planned to support pilots, captains and navigating

officers when manoeuvring a ship in narrow waters,

moreover it also may be applied for the improve-

ment of the on board collision avoidance process.

There are ongoing investigations to enhance the ex-

isting collision warnings by using predicted

manoeuvring characteristic data for adaptation of

alarm thresholds in contrary to conventional

CPA/TCPA calculations and fixed limit values to be

applied to every encounter situation without any dis-

tinction of the prevailing circumstances of the en-

counters

Core element of this new approach is the risk

model for situation assessment (Hilgert, Baldauf

1996) differing between the three types of encounter

situations and additionally taking into account the

two conditions of visibility as laid down in the Inter-

national Rules for Preventing Collisions at Sea as

well. Furthermore the concept is also applied to the

new IMO's definition given in the new performance

standards for INS (IMO, (2007) and allows for in-

troducing situation dependent collision alert catego-

ries "Caution", "Warning" and "Alarm" as well.

Cautions and warnings may be switched off by the

operator, but alarms may not.

For self adaptation of thresholds different CPA

limits are foreseen, which will be set automatically

according to the hydrodynamic safe passing distance

related to the dimensions of the involved ships, the

actual sea area and visibility conditions as well.

As suggestion for initial basic values CPA limits

were determined by detailed field study. To ensure a

wide range of user acceptance one emphasis was

laid on the navigators' behaviour and taken into ac-

count. From the point of view of well experienced

navigators it is rather more practical to determine the

safe passing distance with respect to usual data. Un-

der pragmatic aspects and according to the investiga-

tions performed it can be assumed that the nominal

safe passing distance have to be in relation to the

ship's length of the largest vessel L

max

involved in an

encounter situation (L

max

should not be less than 1

cbl). Taking into account the different types of en-

counter situations as defined in the COLREGs a fac-

tor "f

" is necessary which depends on the kind of

situation "x" and the can be determined by Safe

Passing Distance (nominal) = f

• L

max

The values, given in Table 1, are derived from

several investigations (see i.a. Baldauf (2004)) and

are suggested for the four main types of encounter

situations.

Table 1. Recommendation for basic values to calculate situa-

tion dependent threshold

___________________________________________________

kind of encounter f

situation (good visibility) (restricted visibility)

___________________________________________________

head-on situation 2.5 5

meeting port/port-side

overtaking 2.5 5

head-on situation 5 10

meeting stb/stb

crossing situation 5 10

___________________________________________________

These values were proved by simulation studies

and are valid under the conditions "open sea" for

good (column 2) and restricted visibility (column 3).

These values have to be applied to the actual en-

counter situations.

Table 2. Response times for turning manoeuvre depending on

own ships speed and rudder angles

___________________________________________________

Own Rudder time for course covered

speed angle (°) alteration of distance t

90°

(kt) 90° t

90°

(min) d

(nm)

___________________________________________________

24 hard a- 2:25 min 0,97

starboard

24 starboard 3:51 min 1,54

22 starboard 3:24 min 1,25

22 starboard 4:10 min 1,53

___________________________________________________

With respect to the technical setup for predictions

described above manoeuvring data, especially the

response time for a potential course change of 90°,

may be taken from extracted processed VDR record-

ings and used for automatic adaptation of the TCPA

related limits of the dangerous target alarms, either

by taking them directly from a database or by calcu-

lations using the fast time simulation algorithms.

The response time for turning manoeuvre is a fun-

damental value to avoid a collision. Such response

times are only available to captains on board for

some standard manoeuvres under selected environ-

mental and loading conditions as well and they are

usually neither exactly known nor applicable to the

prevailing circumstances of a concrete dangerous

situation to be solved. A sample of a standard set of

response times for a usual sized 5.000 TEUs con-

tainer vessel is given in Table 2.

As stated before, when applying the drafted con-

cept for situation dependent alarm thresholds those

values should be determined by means of the

manoeuvring prediction module. The principal ap-

plication's structure and the relevant data flows are

given in the figure below.

Figure 7: Principal application structure and data flow for self-

adaptation of thresholds for collision alerts

First studies applying the situation dependent

thresholds for detection of dangerous encounter situ-

ations in overall traffic scenarios in sea areas off the

coast monitored by VTS resulted in a reduction of

the number of collision alerts by 40%.

5 SUMMARY

A prototype software module for an On-line

Manoeuvring Assistance is developed based on a

prediction tool using advanced simulation technolo-

gy on board of ships. The results of rudder and en-

gine control changes will be immediately displayed

in an Electronic chart environment to be used for

manual correcting steering actions. The system was

tested using the excellent resources for research and

development of the Maritime Simulation Centre

Warnemunde and can be used also as a training tool

in student courses. During test trials several

manoeuvring situations were managed with an in-

creased performance when using the prediction tool.

A concept for the application of the tools for purpos-

es of collision avoidance is developed in order to re-

duce the number of alarms.

ACKNOWLEDGEMENTS

The research results presented in this paper were

partly achieved in the research project Condition-

based navigational displays (ZUMANZ) belonging

to the “Maritime Safety Assistance Rostock” consor-

tia funded by the German Federal Ministry of Edu-

cation and Research and surveyed by Research Cen-

tre Jülich as well as under the European MarNIS

project, funded by the European Commission, De-

partment for Energy and Transport.

REFERENCES

Benedict, Kirchhoff, Baldauf, Gluch, Fischer, KO (2007),

Concept for On-board Prediction Displays based on actual

Ship Condition and Manoeuvring Simulation for Naviga-

tion and Shiphandling. Paper at IMSF Conference 2007,

August 20 – 24 2007 Busan / Korea

Baldauf, M.; Benedict, K.; Fischer, S.; Gluch, M.;, Kirchhoff,

M.: DE (2007) Navigation Displays for Ships with simula-

tion based Predictions. (German) in Rötting et al (Eds.):

Prospektive Gestaltung von Mensch-Technik-Interaktion.

Fortschr.-Ber. VDI Reihe 22 Nr. 25, Düsseldorf: VDI-

Verlag 2007, ISBN 978-3-18-302522-0

Baldauf, Benedict, Wilske, Grundevik, Klepsvik, POL (2007),

Combination of Navigational Equipment and VDR Based

Information to Enhance Alert Management. in Weintrit, A.

(Ed.): Advances in Marine Navigation and Safety of Trans-

portation. Monograph Gdynia Maritime University (Po-

land) and The Nautical Institute London (UK), Gdynia,

June 2007, ISBN 978-83-7421-018-8

Baldauf, M. (2004), Enhanced Warning Functions for on board

Collision Avoidance using AIS and VDR information, In:

R. Dauer, A. de la Pena, J. Puig (eds): International Con-

gress on Maritime Technological Innovations and Research

- Proceedings. SCI UPC, Barcelona, ISBN 84-7653-861-8

Oltmann, JP (2003), Identification of Hydrodynamic Damping

Derivatives – a Pragmatic Approach, MARSIM - Interna-

tional Conference on Marine Simulation and Ship Manoeu-

vrability, Kanazawa, Japan, August 25th – 28th

Benedict, Baldauf, Felsenstein, Kirchhoff, JP (2003), Comput-

er-based support for the evaluation of ship handling simula-

tor exercise results, MARSIM - International Conference

on Marine Simulation and Ship Manoeuvrability, Kanaza-

wa, Japan, August 25th – 28th

Benedict, US (2000), Integrated Operation of Bridge-, Engine

Room- and VTS-Simulators in the Maritime Simulation

Centre Warnemünde, Conference on Simulation CAORF /

JSACC 2000, New York, 3-7 July 2000, Proceedings Vol. 1

Benedict, K. et al: (2006) Combining Fast-Time Simulation

and Automatic Assessment for Tuning of Simulator Ship

Models. MARSIM - International Conference on Marine

Simulation and Ship Manoeuvrability, Terschelling, Neth-

erlands, June 25th – 30th Proceedings, M-Paper 19 p. 1-9

Clarke, Horn, UK (1997), Estimation of Hydrodynamic De-

rivatives, Proceedings of the 11th Ship Control Systems

Symposium, Southampton, Vol. 3, pp. 275-289

Hilgert, H.; Baldauf, M. (1996), A common risk model for the

assessment of encounter situations onboard ships. in: Mari-

time Collision and Prevention, Chiavari Publishing, Surrey,

England. 1996, ISBN 0952059223

Clarke, Gedling, Hine, UK (1983), The Application of

Manoeuvring Criteria in Hull Design Using Linear Theory,

Transactions of the RINA, London,

IMO 2007. Revised performance standards for integrated navi-

gation systems (INS). MSC.252(83). London: IMO