725
1 INTRODUCTION
Theideaofunmannedtransportvehiclesisnotnew.
At each stage of development, it should take into
account vehicle safety and the safety of the
environment (including other users). First
implementationsofunmannedvehiclestookplacein
space transport. Then, commercial use of rail
transport was implemented
to carry freight and
people.Furthercommercialapplicationstakeplacein
airtransportformilitaryandciviltransportpurposes.
Commercialuseinroadtransportusingautonomous
vehicles are known to be realized within relatively
smallareas,inaccessibleforotherusers.Commonuse
of autonomous cars and trucks on generally
accessed
roads are still at the phase of research.
Similaractionsaretakeninwatertransport.
Watertransportofpeopleandgoodsisoneofthe
oldest known methods used by humans on local
routesandforverylargedistances.Todate,thereare
no known commercial solutions for the operation of
autonomous ships. However, tests are in progress
involvingautonomousshipsinselectedareasoflocal
importance.
Theoperationofmaritimeautonomousshipscalls
forworkingoutinternationalsolutionstolegalissues.
Therefore, the International Maritime Organisation
hasundertakentocoordinaterelevantwork.Alively
discussion is conducted at meetings of
IMOʹs
committees and subcommittees with large
involvement of Member States, affiliated
organisations and commercial companies, which see
their opportunities to become leaders in
implementing technologies for autonomous ship
operationsatsea.
The implementation of technologies for maritime
autonomous ships, trading locally or across the
oceans, must be preceded by solving
various
problems.Theserefertobothlegalandtechnological
requirementsrelatingtoshipcontrolandsupervision,
andotherissues.Theproblemstobetackledinclude
the maintenance of autonomous ships to ensure,
similarlytoaviation,theirfailure‐freework,problems
related to property salvage, prevention of
environmentaldisasters, threatsrelated
tocrime and
cyber crime. Apart from these problems of global
safety,thehumanfactorremainsessential:trainingof
thecrewtoworkintheperiodoftransitionandlater,
Operations of Maritime Autonomous Surface Ships
Z.Pietrzykowski&J.Hajduk
M
aritimeUniversityofSzczecin,Szczecin,Poland
ABSTRACT:Advancingtechnologiescreateuniqueopportunitiesforconstructingautonomousships,which,in
turn,raisegrowinginterestofthemaritimeindustry,shipownersinparticular.Theseauthorshaveanalyzed
actionstakeninthisfieldand some aspectsrelatedto theoperationsofmaritimeautonomoussurface ships
(MASS).
Thepresentedcasestudyreferstoashipwithaskeletoncrewonadeepseavoyage,wheretheship’s
autonomyisnarrowedtothefourthstageoftransporttask–seavoyageanditsnavigationalaspect.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 13
Number 4
December 2019
DOI:10.12716/1001.13.04.04
726
and personnel supervising ships from land‐based
operational centers. Technological changes should
also address typical issues that ships with crew on
board normally solve through their established
procedures, e.g. berthing and mooring of an
unmanned ship. It seems that solving the above
problems will take time, having rather evolutionary
than
revolutionary character. The evolution is
expectedtotakeyears,asthetechnicalsolutionswill
undergo multiple tests, with skeleton crews still on
board. Thesetestswillbestrictlyverified,especially
by the demanding maritime community. Therefore,
thecurrentdiscussionfocuseson theintroductionof
new technological solutions, parallel tothe
modification of ship equipment used so far, and
limiting current personnel in favour of subsequent
risinglevelsofautonomy.
2 ACTIONSFORMASSIMPLEMENTATION
Trialsofautonomousshipoperationontheseasand
oceanshavealreadybeendone.Atthecurrentstage,
attempts are made to reduce, not eliminate, the
human
factorfromworkonboard.Atthesametime,
newtechnologicalsolutionsallownotonlyasafeship
movement without qualified crew on the navigating
bridge, but also for the conduct of sea trials of
unmannedshipsinconfinedwaters. The pressure is
growing from the maritime industry, including
shipowners,
onconsultative,legislativeandexecutive
bodiestodevelopandimplement solutionsallowing
fortheoperationofautonomousships,includingthe
InternationalMaritimeOrganisation.
TheIMOhasbeenestablishedtosolveproblemsof
the marine environment safety and protection. Its
forumisopentodiscussnotonlynewideasrelatedto
challenges
of technological development, changing
the existing order and established procedures. The
organisationhastoworkout regulatory instruments
allowingtheuseofnewtechnologies,especiallythose
addressing the replacement of people by intelligent
systems.ThepreviousinformaldiscussionsonMASS
have been finally included in the IMOʹs working
schedule
andsince2017hasbeen on the agendas of
committee and subcommittee meetings. After the
publicationofthedocuments[1, 2] the reportof the
98.SessionoftheMaritimeSafetyCommittee[3],in
the point defining the scope of MSCʹs work
embraced the need to identify the problem and
undertakeascopingexerciseforMASS.Thatseemsto
beabreakthroughmoment.Acorrespondencegroup
was established [24], while at 99. MSC Session
working groups were also set up [23, 36]. Many
countries and IMO affiliated organizations have
considered the development of new concepts and
solutionssurroundingtheMASS
astheirduty[5,6,7,
8,9,10,11,12,13,14,15,16,17,18,19,21,25,26,27,
29, 30, 31, 32, 34, 35]. IMOʹs Secretariat is now
attempting to consolidate these proposals and
concepts[28,33].Besides,reportsofthetestscarried
outhavebeen
published[20,22].
TosumuptheproposedMASSsolutions,itshould
be underlined that the majority of States and
insitutions understand the many challenges to be
faced.Atthisstage,theydonotfocusontechnology,
whose development progresses in its own right. It
seems crucial to look into
the legal aspect of
internationalshippingand makeefforts,as
technologyisrapidlyadvancing,toadjustregulatory
instruments and avoid a situation where
technological progress is slowed down or the law is
simply bypassed. Most of the proposals take into
account the evolution of MASS technology, and a
numberofproposals
focusonmodifyingtheexisting
international legislation, where various levels of
MASSautonomywouldbedefined.Thisparticularly
refers to the human factor. To put it simple, the
problem to be solved is to determine levels of
autonomyandproposetechnologicalsolutions,which
willenabletheradicalreductionofshipʹ
spersonnelin
thefirstplace.Today,itisunderstandableandlogical.
Further steps related to the exclusion of traditional
crew on a sea‐going ship, transfering the ship
movement control to a land‐based centreand
finally, regarding a sea‐going ship as moving
autonomously, today are difficult to implement
as
wellastocomprehendsuchconceptofshipping.
Atpresent,asdefinedin[36],theautonomousship
isunderstood“asashipwhich,toavaryingdegree,
can operate independent of human interaction”.
Various divisions into levels of autonomy are
proposed,whichlargelyresultsfromthecomplexity
ofthe
issue.Theproblemisvital,though,asitdirectly
translates into engineering and technological
solutions, and therefore, on ship operation
understoodasitsuseandmaintenance.Anotherissue
isthewayandscope of executingeachphaseofthe
transporttask‐carriageofcargoandpeople:
1 arrivalattheplace
ofloading,
2 shipʹspreparationforloading,
3 cargoloading,
4 seapassage,
5 dischargeofcargo.
Thesephasesmaybecarriedoutatvaryinglevels
ofshipʹsautonomy,frommanualtofullyautonomous
modes. From this perspective, the levels of shipʹs
autonomymayrefertothe
environmentinwhichthe
shipoperates,wheredistinctareasinclude[37]:fleet
management and the control and supervision of
vessel traffic, organization and execution of port
operations,organizationandmonitoringoftransport
processes(logistics,transport,forwarding)(Fig.1).
Figure1.Environmentofthemaritimeautonomoussurface
ship.Source:[37]
Consideringtheoperationofa conventionalship,
wecanpointoutsuchmaintasksoftheshipas:
727
navigation;
steering;
communication;
propulsion;
power;
cargohandling;
passengerservice,
maintenanceandrepairwork.
For autonomous ships, operating independent of
human interactionrequires the development of
technological solutions for them and setting forth
standards/ requirements concerning safety, security,
reliabilityandefficiency.
Technologicalsolutions
arealreadyavailabletoa
largeextent.Intensiveresearchiscontinuedintoother
developments, including simulation tests, tests of
physical models and the construction of prototypes.
The research is often done as part of national or
internationalprojects,e.g.:
MUNN – Maritime Unmanned Navigation
through Intelligence in Networks,
http://www.unmanned‐
ship.org/munin/
AAWA – Advanced Autonomous Waterborne
Applications Initiative, https://www.rolls‐
royce.com/~/media/Files/R /Rolls‐
Royce/documents/customers/marine /ship‐
intel/aawa‐whitepaper‐210616.pdf
STM Validation – Sea Traffic Management
ValidationProject,http://stmvalidation.eu/
AVAL – Autonomous Vessel with an Air Look,
http://aval‐project.pl/
Testing technological solutions in real conditions
requires the identification of places where tests
may
beperformedandtheprinciples toensure thesafety
ofothershipsproceedinginthevicinity.Theworkis
already ongoing and mainly involves the
administrations of maritime countries, as well as a
recently established MSC working group on MASS
[10].
Putting autonomous ships into service calls for
appropriate
regulations, the basis for activities of
classification societies and insurance companies.
LegislativeworkisconductedbytheIMOʹsMaritime
SafetyCommittee.Itisassumedthatchangeswillbe
evolutionary,startingfroman analysisand
modification of the existing regulations referring to
maritimesafetyandsecurity.Ithasbeenproposedto
prioritize a review of IMO mandatory instruments.
First, anreviewoftheexistingIMOregulationswill
be carried out to determine their applicability to
MASS: A) apply to MASS and prevent MASS
operations; B) apply to MASS and do not prevent
MASSoperationsandrequirenoactions;C)applyto
MASSanddonotpreventMASSoperationsbutmay
needtobeamendedorclarified,and/ormaycontain
gaps;D)havenoapplicationtoMASSoperations.
The next step will consist in determining the
appropriate route to assess the operation of MASS,
taking into account, inter alia, the human factor,
technological
andoperationalfactorsby:1)adopting
theequivalenceprovidedforintheregulations orthe
possibility to extend interpretation; 2) changing
existinginstruments;3)developingnewinstruments;
4)rejectingthesaidsolutions1to3.Theresultsofthe
aboveactionswill giverisetochanges, supplements
and will introduce
additionalregulations on the
constructionandoperationofMASS.
Today, considerations concerning means of
transport,includingships,takeintoaccounttheirlife
cycle:from demandtoscrapping. Thus,definingthe
conditions of operation and equipment of
autonomous ships we have to take into account all
stakeholderspresentinthatcycle.Five
groupsofsuch
entities can be distinguished. These are shipowners
(1),industry (2),researchand development
institutions(3),classificationsocietiesandinsurance
firms(4)andmaritimeadministrations(5).Theiraims
are different. The first four groups are business
organisations striving to achieve economic results.
Maritime administrations function as coordinators
balancing
the targets of the other stakeholders,
approvingandsupervisingtheentireprocess,caring
for the environment, including its social, economic
andpoliticalaspects(Figure2).
a)b)
Figure2.StakeholdersintheMASSlifecycle:a)atpresent;
b)expected
Insimplifiedtermswecansay that these entities
attemptto:
shipowners:minimizethecostsofshipequipment
andoperation,includingcrewcosts,
the industry: maximize profits by using new
technologies
R&D institutions: develop cognitive functions by
developingnewtechnologies
classification societies and insurance forms:
minimize risk
assessment errors for preventive
purposesintheprocessofoperation,
maritime administrations: formalize standards to
minimizeoperationalrisksapplyingtheʹaslowas
reasonablypracticableʹ(ALARP)principle.
3 EQUIPMENTOFTHEAUTONOMOUSSHIP
Current considerations of the concept of developing
autonomous shipsare focused on finding a
consensusat
aninternationallevel,sothatstateswill
beable torecognizethestatusofautonomousships.
In the short run, such ships could be manned by
fewerqualifiedcrewmembers,whilein the long run
the unmanned ships would be remotely supervised
by land‐based operators. The discussion so far has
been
concentrated on setting up international
regulations that would define the shipʹs equipment
duringthetransitionperiodinthecontextofcurrently
bindingregulations.Technologicaldevelopmentsthat
728
spurtheprogressofdiscussionsarenotlimitedinthe
sense of specific solutions or directions of further
advancementstobeadoptedonocean‐goingvessels.
Atpresent,theshipboardnavigational,life‐saving
and radio equipment of a marine ship is strictly
defined by the SOLAS Convention. The level of
automationusedisoneofthecriteriaindetermining
the minimum manning of a sea ‐going ship. The
installation of additional systems supporting
navigatorʹsorengineerʹsworkdoesnotdirectlyaffect
the composition of shipʹs personnel. Behind the
current philosophy of fitting the ship with various
technological
novelties, such as decision support
systems,is the intention of limiting human errors
and reducing the number of accidents at sea.
However, such additional ship equipment, if not
required by law, is not common due to economical
reasons. Each investment in additional equipment
should be economically viable. It is only a
legal
consensus at international level that will lead to a
situation where investing in IT equipment on a sea‐
goingshipwillbeeconomicallyjustified,i.e.itwillbe
possible to reduce the number of qualified crew
members. This is the reason of the pressure from IT
equipmentmakersas
wellasshipownerswhointhe
long run expect, rather than fewer accidents at sea,
winningtechnologicaladvantageresultinginlower
operatingcosts,whichiscompetitiveadvantage.
There are factors affecting the fitting of the
autonomousship.Theseare:
manning(without/withcrew),
theareaofautonomy(stages
ofthetransporttask).
thescopeofautonomy(navigation,steering,…).
Besides,foralltheabovevariants,instandardand
emergency situations it should be a possibility of
taking over the control of the ship by an onboard
operator or remotely by a land‐based operator. In
addition,in
caseofocean shipping inparticular,the
problem of maintenance remains ( during operation
orwhiledocked).
It is assumed that in the initial period, small
unmanned ships in coastal shipping and sea‐going
shipswithaskeletoncrewwillbeputintooperation.
In the former case, e.g. on small
passenger ferries,
autonomywillbeappliedtoallstagesofthetransport
task.Inthelattercase,itisassumedthatseavoyage
willbeexecutedinanautonomousmanner, whilethe
other parts of the task will be carried out by an
operator on board or at the land‐
based centre or
operators sent to the ship for executing scheduled
operations,e.g.mooring.Giventheaboveconditions,
itseemsthat the equipmentoftheautonomousship
willbeadaptedtotheestablishedmanning,area and
scope of its autonomy. The reference point may be
standard ship equipment as required by
relevant
conventions. However, one can expect that new
technologicalsolutionswillbeintroduced,thatisnew
systemsandequipmentonshipsandinland‐based
centres. This entails the need to define a
template/templates specifying technical
measures/technologiesfortheassumedmanning,area
and autonomy range of the ship. It is
necessary to
ensure appropriate infrastructure on land, including
theequipment ofland‐basedcentres,performing the
supervisory functions and being able to take over
remotecontrolof theship.A preliminaryconceptof
suchatemplateisproposedin[8].
In the next step, technological solutions will be
describedindetail,
including,forexample,theMASS
decision‐making module. Particular solutions may
compriseexistingdecisionsupportsystems,e.g.those
adddressing collision avoidance.Thechallenges also
refer to guaranteeing an appropriate level of data
quality, system security and system reliability.
Implementationofnewsolutionsshouldbepreceded
byacomprehensiveriskassessment
(IMO‐FSA).An
exampleinthiscasemaybetheequipmentofaship
with a skeleton crew on a deep‐sea voyage, with
shipʹs autonomy restricted to the fourth stage of
transporttask – the navigational aspect of sea
voyage(seeChapter1).
4 MANNINGOFANAUTONOMOUSSHIP

The human as an operator of a complex transport
system,inparticularinmaritimetransportsector,isa
complexissue.Ifwelook intothecurrentstudiesand
proposals concerning the timetable of work and the
overall creation of the conceptsof maritime
autonomous ships, four states related to the
ship
manningcanbedistinguished[10]:
Qualifiedpersonnel,inaccordancewiththeSTCW
Convention, are on board. The number of crew
members is reduced compared to conventional
ships.
Degree one: Ship with automated processes and
decisionsupport:Seafarersareonboardtooperate
and control shipboard systems and functions.
Some operations may be automated and at times
beunsupervisedbutwithseafarersonboardready
totakecontrol.
Degree two: Remotely controlled ship with
seafarers on board: The ship is controlled and
operated from another location. Seafarers are
availableonboardto take control and to operate
the
shipboardsystemsandfunctions.
Therearenoqualifiedseafarersonboard,ableto
takeovershipcontrol,
Degree three: Remotely controlled ship without
seafarers on board: The ship is controlled and
operated from another location. There are no
seafarersonboard.
Degreefour:Fullyautonomousship:
Theoperating
system of the ship is able to make decisions and
determineactionsbyitself.
Thefirstcaserequiresthatthenumberoftrained
seafarers on board complies with the currently
applicable STCW Convention. The problem to be
sorted out is the process and scale of reducing a
conventional crew,
to match shipʹs degree of
autonomy. This is a relatively simple issue, which
requires certain legislative changes, but it is
intuitivelyunderstandable.Thesecondissuerefersto
hiring a considerably smaller number of personnel
withappropriateSTCWcertificatesand,ifnecessary,
staff required for running maintenance of shipboard
systems.
The problem concerns the training
requirementsforsuchpersonnelinmarinerescueand
life‐saving skills. The third, completely new issue,
resultsfromtheeliminationofshipʹsdeckcrewand
729
the need to develop th concept for qualifications of
personnel working in land‐based control and
supervision centres for autonomous ships. The basic
question addresses the scale and scope of required
conventional marine competencies. In other words,
thequestioniswhetherland‐basedpersonnelneedto
haveanymarinecompetencies,
orperhapsonlythose
oftransportsystemoperators.
An interesting proposal for creating a system of
levels of autonomy and control was put forward by
Australia and other countries [9]. The levels of
technicalautonomywereseparatedfromoperational
control,managed by qualified personnel on board.
The concept combines levels
of technical autonomy
withthepresenceor absenceofpersonnelonboard.
Four levels of technical autonomy have been
established:
A0,Manual.Manualoperationandcontrolofship
systemsandfunctions, includingbasicindividual
system level automation for simple tasks and
functions.
A1, Delegated. Permission is required for
the
execution of functions, decisions and actions; the
operatorcanoverridethesystematanystage.
A2, Supervised. The qualified operator is always
informed of all decisions taken by the system.
Permission of the qualified operator is not
requiredfortheshipsystemtoexecutefunctions,
decisions and actions; the
qualified operator can
overridethesystematanystage.
A3, Autonomous. The qualified operator is
informed by the system in case of emergency or
when ship systems are outside of defined
parameters.Permissionofthequalifiedoperatoris
not required for the ship system to execute
functions, decisions and
actions; the qualified
operator can override the ship system when
outside of defined parameters. Provided the
boundaries of the ship system are not exceeded,
ʺhumancontrolʺbecomesʺhumansupervisionʺ.
Operational control of the human distinguishes
onlytwostates:
B0,Noqualifiedoperatorsonboardbutqualified
operatorsavailable
ataremotelocation
B1,Qualifiedoperatorsonboard
Each of the above concepts has its strengths and
weaknesses.Theformer,representedbytheworking
group [10], is a closed concept, with two states
including qualifiedcrew on board and two states
without crew. The latter concept, developed by
Australia
[9], sets forth more options and is more
flexible in the implementation of the concept of
maritimeautonomousships.
An analysis of the two documents [9, 10] allows
for a general conclusion that automation of each
processleadstogreaterrepeatability,enhancementof
safetyandgradualreplacementofpeople,which
will
eventually bring economical profits. This is a key
statement,astheexpectedeconomicbenefitsresulting
fromthegradualreplacementofmanonasea ‐going
vesselshouldbringbetteroperationofthesystemand
improved safety at sea, the outcome of gradual
exclusion of the human factor. This thesis will
be
verified inpractice, through multiple programmes
andtrials,similarlytopublicroadtransport.
In the context of training of seafarers or,in the
future,land‐basedoperatorsofMASSsystems,while
accepting the need to link technological automation
with maritime training, it seems relevant to build
these developments on
existing international
regulations, including the provisions of STCW and
SOLASconventions.
Today, the STCW Convention fully regulates the
training of seafarers, stipulating theoretical training
supportedbymandatory sea service,whichtogether
areaprerequisite tocompetenciesneededinspecific
positions.TheSOLASConventioninRegulationV/14
requires a minimum manning that
is determined,
interalia,bythedegreeofshipautomation.Aslong
as the first and second levels of autonomy are
examined[10]orstateB1[9]thatassumethepresence
ofqualifiedorreduced crew,noconflictarises. An
international consensus is needed in working out
morepreciselinks
betweenthedegreeofautomation
and the number of qualified personnel required on
board.
Itisdifferentwhenitcomestoimplementingthe
thirdandfourthlevelsofautonomy[10]orstateB1
[9]. The absence of qualified personnel on the ship
results in an entirelynew operational and legal
situation.
In the field of trainingland‐based
operators/supervisorsofaMASSsystem,therewill
be doubts as to at which particular moment
supervisionshouldbeabortedandremoteoperation
started. In the initial period, additional training of
navigating officers can take place, who, with extra
qualifications, may be
employedas land operators
of autonomous ships. In the long run, however,
assumingsignificantreductionsorcompleteexclusion
byautomaticsystems,itseemsnowisthetimetostart
work on setting instructional framework for such
personnel.Undoubtedly, theoreticaltrainingwill not
facemanyproblems.Thequestionconcernspractice.
Today, without
understanding the size and mass of
the ship, especially the hull behaviour in stormy
waves, it is difficult to imagine correct decision
making. Undoubtedly, to better understand the
problems of shipʹs work in waves,enhanced by
climaticchanges,gainingexperienceontrainingships
shouldnotbeexcluded,whichwill
giveatleastbasic
understandingofshipʹsbehaviourinheavyweather.
Theexamineddocumentsdonotclearlyformulate
the situation where qualified crew is not present on
the ship, but personnel performing servicing and
maintenanceworkare.Itishardtoimaginetodaythat
autonomousshipswithoutqualifiedseafarerswill
be
servicedonlyduringshortstaysinport.Itisregarded
as an ultimate target such as already effectively
functioninginaviation.Themodelcurrentlyoperated
in maritime shipping is just opposite. The normal
practiceisthat shipscarryspare parts forshipboard
equipment,operationalmaterials(e.g.lubricatingoil)
and
qualifiedpersonnelabletoperformevencomplex
repairs.Servicesofferedashoreareusedoccasionally
or in major failure situations. As the marine
environmentisveryaggressive,itishardtoimagine
thatregularmaintenanceandrepairworkenrouteis
given up, and done only in ports, along with an
increase
inthefrequencyofclasssurveysperformed
inshiprepairyards(Figure3).
730
Figure3.On‐shipoperationaltasks
Undoubtedly, the experience of the ongoing
operation of ships indicates in such situationsthe
need to increase the number of salvage ships that
even in the long run will not be converted into
autonomousshipswithoutqualifiedpersonnel.
5 CONTROLANDSUPERVISIONOF
AUTONOMOUSSHIPS
Todayland‐basedsafetyrelatedassistance
toshipsat
sea has a few aspects. Each of the currently
consideredaspectsisbasedonthesheerfactthaton
board are qualified seafarers who utilize various
types of information worked out on land. The basic
functionofsuchcentresorarrangementsistoensure
an adequate level
of safety of shipping in standard
andemergencysituations.
Weather services have been provided on the
oceans for years, tasked to recommend optimal
weather routes ensuring safe navigation within
optimizedperiodoftime.Thesearepaid‐upservices.
Shipowners pay for the access to information, and
ships usually get specific recommendations,
which
may be taken into account by the captain fully,
partiallyorignored.Forfutureautonomousships,the
roleofsuchserviceswillundoubtedlybeincreased.In
thecaseofunmannedvessels,arecommendedocean
routewillchangeitsstatustothemandatoryroute,as
it seems that a system
operator will not be able to
ignore such recommendations. Naturally, working
outanoptimalweatherrouteisalsopossibleonboard
using autonomous software. At present this process
requiresthepresenceofaqualifiedpersononboard.
In frequented coastal routes and port approaches
VesselTrafficServicesystemsareinuse.
Theroleof
such systems for unmanned ships passing near a
coastalstateandcallingataspecificportwillhaveto
be verified and extended. This applies to automatic
reporting and responses to emergency situations
occurringinareascoveredbyVTSsystems.
In extraordinary situations at sea,including
emergencies,
presently the master is assisted by a
designated person or an entire emergency response
team. The essential feature of such assistance to the
master is that the designated personnel have sea
experience and seamanship, usually supported by a
representative of the insurer. The captain receives
current advice on actions to be
taken, while the
emergency team arrange for rescue party, port of
shelter and undertakes other external actions. In an
emergency situation occurring on a unmanned ship,
theroleofsuchateammustbecoordinatedwiththe
shipoperatorattheland‐basedfacility.
The operation of unmanned autonomous ships
will require establishment of shipowner centres for
shipmanagement,regardlessoftheautonomylevel3
[10]ofremotelycontrolledshipsorlevel4[10]offully
autonomous ships. Monitoring function will be
performed in these cases. When such concept is
implemented, a group of operators controlling
autonomous ships and a monitoring
group will be
established. In the long run, the STCW Convention
might be broadened to include sections on the
training process and competencies of land‐based
operatorsatthethirdandfourthlevelofautonomy.
6 THEOPERATIONOFTHEAUTONOMOUSSHIP
‐CASESTUDY
The study focuses on the equipment
of a sea‐going
shipwithskeletoncrew,withitsautonomynarrowed
to the fourth stage of the transport task‐shipʹs
navigationonaseavoyage(comparechapters1and
2). Proposed guidelines for autonomous vehicles:
ships[Korea] and cars / trucks[NHTSA] have
been taken into account. The
following method of
analysisisadopted:identificationoftheselectedstage
tasks, determination of the necessary functionalities
and implementation. The following tasks are
distinguished:
acquisition, integration and fusion of data
(situationalawareness),
avoidanceofcollisions andgroundings(analysis
and assessment of the navigational situation,
determination of a safe
trajectory in a collision
situationorriskofgrounding),
navigation along a set trajectory according to the
voyage plan, and in case of a collision situation,
alongasafedeterminedtrajectory,
human‐machine interface (HMI), ship‐shore
communication and alertingin emergency
situations(emergencyprocedures),
remotemonitoring/control.
The above tasks have been considered from
variouspointsofview:
Operational Design Domain (ODD): operating
parameters, environmental conditions, and any
otherdomainconstraints,
situational awareness and collision avoidance –
ObjectAndEventDetectionAndResponseOEDR:
detection and respond adequately to
circumstances,
Human Machine Interface HMI –
means and
eventscalling/generatingtheinteractionseafarer–
MASS,
Minimum Risk Condition MRC – fallback:
transition to a condition with minimal risk: risk
control options or transition from autonomous
modetohumancontrol,
protection from intrusions and disruption –
cybersecurity: incorporation of cybersecurity into
thedesignofMASS.
Acquisition, integration and fusion of data
(situational awareness). The MASS operation is
based on information about own ship and the
environment, obtained from devices and systems
731
availableonboardinanyconditions.Thedefinitionof
theconditionsforʹproperlookoutʹforaMASSseems
to be of key importance. This means the need to
indicateadditional,comparedtoconventionalships,
dataandsystemsandequipmentfordataacquisition.
Similarly to conventional ships, mandatory systems,
scope, accuracy
and reliability of data after the
processesofintegrationandfusionofdatashouldbe
defined.Morecompleteuseofalreadyavailabledata
may be a partial alternative. The aim is to provide
dataneededfortheperformanceofMASStasks and
the supervision by ship crew or the land‐
based
operators. The minimum risk condition refers to
situations where the correct operation of the ship is
threatened. It is related to ensuring appropriate
redundancy of systems and equipment for the
acquisition, integration and fusion of data in case
those in use suffer a failure, and to assure data
reliability and
accuracy through appropriate
methods and tools of cybersecurity (standards and
levels).
Navigationalongasettrajectoryaccordingtothe
voyageplan,andinthecaseofacollisionsituation,
alongasafedeterminedtrajectory.Theprovisionof
safeoperatingconditionsinvolvesdefiningthelimits
ofshipmovementparameters,takinginto
accountthe
shipʹs manoeuvring capabilities in current
environmental parameters(hydrometeorological
parameters, other ships, navigational obstructions,
etc.).Depending onthe natureofchanges, theymay
resultinactivationofthetaskofcollisionavoidance.
Thequestiontobeansweredinthisrespectisinwhat
conditions and how the crew
or operator should be
informedofsuchevents:shouldanyorspecificevent
besignalled,shoulditbeinformationorwarning,or
instruction to take over the ship control (HMI). The
minimum risk conditions may relate to additional
functions of the autopilot for keeping the ship on a
chosen route. Cybersecurity
should be ensured by
appropriatesolutionsincorporatedinthetaskofdata
acquisitionandintegration.
Avoiding a collision or grounding includes an
analysisandassessmentofthenavigationalsituation,
then determination of a safe trajectory in a collision
situation.Inthiscase,theconditionsofoperationrefer
tocriteriaand
limitationsinthedeterminationofsafe
trajectoryandmethodsfordeterminingsolutions.The
relevant requirements are defined in the existing
navigational decision support systems. Their
solutionsmaybeusedintheconstructionofHMIfor
MASS. The essential functions within risk
minimization will be systems of automatic
communication and negotiations. These
will
contribute to increased situational awareness, in
particular the planned actions of other ships. They
will allow for agreeing on manoeuvres of the
encountering ships, and in a close quarter situation
for coordinating actions to avoid a collision. Like in
thetaskofdataacquisitionandintegration,itwillbe
important to ensure an appropriate level of
cybersecurity.Partoftheadoptedsolutionsmayalso
beusedinthistask.
Communication with the crew and land‐based
centreistoensurethatthecontinuousmonitoringof
MASSvoyageiscarriedoutbythecreworthecentre
operators,withthecapability
oftakingovertheship
control. The loss of MASS communication with the
crewandland‐basedcentreshould be considered as
critical for the safety of navigation. Therefore, the
solutions applied should ensure the communication
in emergencies as well, by using dedicated HMI
designed for variousworking modes (standard,
emergency,etc.).Theminimumriskconditioncanbe
ensured, inter alia, by redundant systems of
communication and synchonized use of additional
meansofcommunicationandsignalling:voiceand/or
light. The MASS‐crew and MASS‐centre
communications must meet adopted standards and
levelsofcybersecurity.
Remote monitoring/control. The implementation
ofthis
taskrequiresensuringcontinuousmonitoring
of the MASS voyage, and control of ship‐based
systems and equipment by land‐based operators in
any conditions, and the possibility of relaying the
control function to the crew through established
procedures. Land‐based centre HMIs should feature
functions,inadditiontothoseinuse,
relatedtothe
monitoring and management of vessel traffic
monitoring and management, and functionalities for
HMIusedbythecrew.Theminimumriskcondition
can be ensured, inter alia, by redundant
communication systems. This task in particular
requires high level of cybersecurity. The need
therefore arises to set forth standards
and levels of
cybersecurityfor MASS‐land‐basedcentre
communicationsystems.
7 SUMMARY
The presented analysis of the operation of the
autonomous ship refers to the equipment of a ship
withaskeletoncrewinthesea‐going voyage, where
shipʹsautonomyisrestrictedtothefourthstageofthe
transporttask,navigationduringtheseavoyage,and
doesnot exhaust the subject of autonomous ship
equipment. It may serve as a template for
consideration of other cases varying in the type of
navigation, ship type or conditions of operation. It
seems inevitable to engage all five groups of
stakeholders
in the process of establishing MASS
operation conditions. The implementation of MASS
technology in maritime navigation is largely
dependent on the activity of the widely understood
maritimeadministration.
ACKNOWLEDGMENTS
Thisresearchoutcomehasbeenachievedundertheresearch
project No 1/S/ITM/ 2016 financed from a subsidy of the
Polish Ministry of Science and Higher Education for
statutoryactivitiesoftheMaritimeUniversityofSzczecin.
REFERENCES
[1]MSC 98/20/2. Maritime Autonomous Surface Ships.
Proposalforaregulatoryscopingexercise.Submittedby
Denmark, Estonia, Finland, Japan, the Netherlands,
Norway, the Republic of Korea, the United Kingdom
andtheUnitedStates.27February2017
732
[2]MSC 98/20/13. Maritime Autonomous Surface Ships.
Proposal for a regulatory scoping exercise. Comments
on MSC 98/20/2. Submitted by the International
TransportWorkersʹFederation(ITF).13April2017
[3]MSC 98/23. REPORT OF THE MARITIME SAFETY
COMMITTEE ONITSNINETY‐EIGHTH SESSION.. 28
June2017
[4]MSC 99/5. REGULATORY SCOPING EXERCISE
FOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS (MASS). Comments on the regulatory scoping
exercise.NotebytheSecretariat.13March2018.
[5]MSC 99/5/1. REGULATORYSCOPING EXERCISE FOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS (MASS). Comments and proposals on the way
forward for the regulatory scopingexercise. Submitted
byIFSMAandITF.22February2018.
[6]MSC 99/5/2. REGULATORYSCOPING EXERCISE FOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS(MASS).Proposalsforthedevelopmentofawork
plan.SubmittedbyICS.8March2018.
[7]MSC 99/5/3. REGULATORYSCOPING EXERCISE FOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS (MASS).
Recommendations on identification of
potential amendments to existing IMO instruments.
Submitted byFinland, Liberia,Singapore, South Africa
andSweden.8March2018.
[8]MSC 99/5/4. REGULATORYSCOPING EXERCISE FOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS. Considerations on and proposals for the
methodology to use within the framework of the
regulatory scoping exercise. Submitted by France. 9
March2018.
[9]MSC 99/5/5. REGULATORYSCOPING EXERCISE FOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS (MASS). Plan of approach for the scoping
exercise. Submitted by Australia, Canada, Denmark,
Estonia, Finland, Japan, the Netherlands, Norway,
Singapore, Sweden, the United Kingdom, the United
States,IMarEST
andIMCA.12March2018.
[10]MSC99/5/6.REGULATORYSCOPINGEXERCISEFOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS(MASS).Considerationsondefinitionsforlevels
and concepts of autonomy. Submitted by Finland. 12
March2018.
[11]MSC99/5/7.REGULATORYSCOPINGEXERCISEFOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS (MASS).
Proposal on the work plan of the
regulatory scoping exercise for the use of MASS.
SubmittedbyChinaandFinland.13March2018.
[12]MSC99/5/8.REGULATORYSCOPINGEXERCISEFOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS(MASS).Recommendationsoncategorizationand
regulatory scoping exercise of MASS. Submitted by
Chinaand
Liberia.13March2018.
[13]MSC99/5/9.REGULATORYSCOPINGEXERCISEFOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS (MASS). Japanʹs perspective on regulatory
scoping exercise for the use of MASS. Submitted by
Japan.13March2018.
[14]MSC 99/5/10. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE
SHIPS(MASS).Generalcommentsonaway
forward.SubmittedbyITF.26March2018.
[15]MSC 99/5/11. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Comments on documents
MSC 99/5, MSC 99/5/2, MSC 99/5/5, MSC 99/5/8 and
MSC99/5/9.SubmittedbyTurkey.27March2018.
[16]MSC
99/5/12. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Comments on document
MSC 99/5/5.Submittedby the United States.27 March
2018.
[17]MSC 99/INF.3. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Final Report: Analysis of
Regulatory Barriers to the use
of Autonomous Ships.
SubmittedbyDenmark.18January2018.
[18]MSC 99/INF.5. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS(MASS).Report of a surveyon what
maritimeprofessionals think aboutautonomous
shipping. Submitted by IFSMA and ITF. 9 February
2018.
[19]MSC 99/INF.8. REGULATORY SCOPING EXERCISE
FOR
THE USE OF MARITIME AUTONOMOUS
SURFACESHIPS(MASS).WorkconductedbytheCMI
International Working Group on Unmanned ships.
SubmittedbyCMI.13February2018.
[20]MSC 99/INF.13. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACESHIPS(MASS).Establishinginternationaltest
areaʺJaakonmeriʺforautonomousvessels.Submittedby
Finland.12March2018.
[21]MSC 99/INF.14. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Studies conducted in Japan
on mandatory regulations relating to Maritime
Autonomous Surface Ships – SOLAS, STCW and
COLREGs.SubmittedbyJapan.13March2018.
[22]MSC 99/INF.16. REGULATORY SCOPING EXERCISE
FOR THE
USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Presentation by Norway on
21 May 2018 on theʺYARA Birkelandʺ development.
SubmittedbyNorway.13March2018.
[23]MSC 99/WP.9. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Report of the Working
Group.23May2018.
[24]MSC
100/5.REGULATORYSCOPINGEXERCISEFOR
THE USE OF MARITIME AUTONOMOUS SURFACE
SHIPS(MASS).ReportoftheCorrespondenceGroupon
MASS.SubmittedbyFinland.28September2018.
[25]MSC 100/5/1. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Proposal for a classification
schemefordegreesofautonomy.Submitted
byISO.31
August2018.
[26]MSC 100/5/2. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACESHIPS(MASS).InterimguidelinesforMASS
trials.SubmittedbyNorwayandBIMCO.28September
2018.
[27]MSC 100/5/3. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Proposals for
the
development of interim guidelines for Maritime
AutonomousSurfaceShips(MASS)trials.Submittedby
RepublicofKorea.28September2018.
[28]MSC 100/5/4. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Comments on document
MSC100/5.NotebytheSecretariat.12October2018.
[29]MSC 100/5/5. REGULATORY
SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Comments on document
MSC100/5.SubmittedbyJapan.11October2018.
[30]MSC 100/5/6. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Comments on document
MSC100/5.Submitted byAustralia,Denmark,Finland,
FranceandTurkey.
12October2018.
[31]MSC 100/5/7. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Comments on document
MSC100/5.SubmittedbyChina.
[32]MSC 100/5/8. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Comments on document
733
MSC100/5.SubmittedbytheUnited States. 12October
2018.
[33]MSC 100/INF.3. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Initial review of IMO
instruments under the purview of MSC. Note by the
Secretariat.9August2018.
[34]MSC 100/INF.6. REGULATORY SCOPING EXERCISE
FOR THE USE
OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Preliminary analysis of the
International Regulations for Preventing Collisions at
Sea,1972.SubmittedbyChina.28September2018.
[35]MSC100/INF.10.REGULATORYSCOPINGEXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Results of technology
assessment on Maritime Autonomous Surface Ships
(MASS).Submitted
byRepublicofKorea.28September
2018.
[36]MSC 100/WP.8. REGULATORY SCOPING EXERCISE
FOR THE USE OF MARITIME AUTONOMOUS
SURFACE SHIPS (MASS). Report of the Working
Group.6December2018.
[37]Pietrzykowski, Z., Decision making in autonomous
shipping‐challenges Autonomous Ship technology
Symposium,Amsterdam2018