Experimental Evaluations on Ship Autonomous Navigation and Collision Avoidance by Intelligent Guidance

Embed Size (px)

DESCRIPTION

Experimental evaluations on autonomous navigationand collision avoidance of ship maneuvers by intelligent guidanceare presented in this paper. These ship maneuvers are conductedon an experimental setup that consists of a navigation and controlplatform and a vessel model, in which the mathematical formulationpresented is actually implemented. The mathematicalformulation of the experimental setup is presented under threemain sections: vessel traffic monitoring and information system,collision avoidance system, and vessel control system. The physicalsystem of the experimental setup is presented under two mainsections: vessel model and navigation and control platform. Thevessel model consists of a scaled ship that has been used in thisstudy. The navigation and control platform has been used to controlthe vessel model and that has been further divided under twosections: hardware structure and software architecture. Therefore,the physical system has been used to conduct ship maneuversin autonomous navigation and collision avoidance experiments.Finally, several collision avoidance situations with two vessels areconsidered in this study. The vessel model is considered as thevessel (i.e., own vessel) thatmakes collision avoidance decisions/actionsand the second vessel (i.e., target vessel) that does not takeany collision avoidance actions is simulated. Finally, successfulexperimental results on several collision avoidance situations withtwo vessels are also presented in this study.

Citation preview

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    IEEE JOURNAL OF OCEANIC ENGINEERING 1

    Experimental Evaluations on Ship AutonomousNavigation and Collision Avoidance

    by Intelligent GuidanceLokukaluge P. Perera, Victor Ferrari, Fernando P. Santos, Miguel A. Hinostroza, and Carlos Guedes Soares

    AbstractExperimental evaluations on autonomous navigationand collision avoidance of ship maneuvers by intelligent guidanceare presented in this paper. These ship maneuvers are conductedon an experimental setup that consists of a navigation and controlplatform and a vessel model, in which the mathematical for-mulation presented is actually implemented. The mathematicalformulation of the experimental setup is presented under threemain sections: vessel traffic monitoring and information system,collision avoidance system, and vessel control system. The physicalsystem of the experimental setup is presented under two mainsections: vessel model and navigation and control platform. Thevessel model consists of a scaled ship that has been used in thisstudy. The navigation and control platform has been used to con-trol the vessel model and that has been further divided under twosections: hardware structure and software architecture. There-fore, the physical system has been used to conduct ship maneuversin autonomous navigation and collision avoidance experiments.Finally, several collision avoidance situations with two vessels areconsidered in this study. The vessel model is considered as thevessel (i.e., own vessel) that makes collision avoidance decisions/ac-tions and the second vessel (i.e., target vessel) that does not takeany collision avoidance actions is simulated. Finally, successfulexperimental results on several collision avoidance situations withtwo vessels are also presented in this study.

    Index TermsCollision avoidance system, decision supportsystem, intelligent guidance, ship collision avoidance, ship colli-sion detection.

    Manuscript received June 24, 2013; revised December 01, 2013; acceptedJanuary 31, 2014. The work of L. P. Perera was supported by the Doctoral Fel-lowship of the Portuguese Foundation for Science and Technology under Con-tract SFRH/BD/46270/2008. This work contributes to the project of Method-ology for ships maneuverability tests with self-propelled models, which is sup-ported by the Portuguese Foundation for Science and Technology under Con-tract PTDC/TRA/74332/2006. This work was presented in part at the 32nd In-ternational Conference on Ocean, Offshore and Arctic Engineering, Nantes,France, June 914, 2013.Associate Editor: K. Takagi.L. Prasad Perera was with the Centre for Marine Technology and Engineering

    (CENTEC), Instituto Superior Tcnico, University of Lisbon, Lisbon 1049-001, Portugal. He is now with Wrtsil Finland Oy, Turku FIN-20811, Finland(e-mail: [email protected]).V. Ferrari is with the Centre for Marine Technology and Engineering

    (CENTEC), Instituto Superior Tcnico, University of Lisbon, Lisbon 1049-001,Portugal and also with the Maritime Research Institute Netherlands (MARIN),Wageningen 7608 PM, The Netherlands (e-mail: [email protected]).F. P. Santos, M. A. Hinostroza, and C. Guedes Soares are with the

    Centre for Marine Technology and Engineering (CENTEC), InstitutoSuperior Tcnico, University of Lisbon, Lisbon 1049-001, Portugal(e-mail: [email protected]; [email protected];[email protected]).Color versions of one or more of the figures in this paper are available online

    at http://ieeexplore.ieee.org.

    I. INTRODUCTION

    A. Maritime Safety

    C ONGESTED sea routes and various offshore operationsenforce ships to make close encounter maneuvers, whichmay lead to some high-risk collision and near-collision situa-tions. However, these issues can be countered by introducingsafety training and safe ship handling procedures in the shippingindustry. Even though these trainings and procedures associatedwith navigators experience could play an important role in safeship navigation, they could also have some limitations due tohuman and economical constrains. Furthermore, even a well-trained and experienced navigator can make wrong navigationjudgments, which can result in ship collisions with human ca-sualties and environmental disasters. For example, even the de-cision-making process of an experienced navigator could be af-fected by unexpected situations with instrumentation and com-munication failures and losing vessel maneuverability condi-tions under various speed and environmental conditions.Therefore, as initiated by e-navigation [1], appropriate navi-

    gation aids should be facilitated to achieve the required safetylevels in the shipping industry. The concept of e-navigation isintroduced by the International Maritime Organization (IMO)and the International Association of Lighthouse Authorities(IALA) [2] for integrating present navigation technologiesand introducing intelligent decision support capabilities tolimit human subjective factors in the shipping industry. Fur-thermore, various studies to create next-generation command,communication, and control platforms that enhance wirelessmonitoring and control functions, including advanced decisionsupport facilities to operate ships remotely under semi or fullyautonomous conditions, have also been proposed [3]. There-fore, the main contribution in this study is also to support theconcept of e-navigation by providing experimental results onship autonomous navigation and collision avoidance based onintelligent guidance as further described in this paper.

    B. Ship Interactions

    In general, ship collision avoidance can be categorized undertwo types of environmental conditions [4]: the coastal phaseand the oceanic phase. The coastal phase consists of collisionavoidance among vessels in confined waters, and the oceanicDigital Object Identifier 10.1109/JOE.2014.2304793

    0364-9059 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    2 IEEE JOURNAL OF OCEANIC ENGINEERING

    phase consists of collision avoidance among vessels in open seaareas. The coastal phase comprises several navigation aids suchas traffic channels for highly dense maritime traffic regions andnavigation guidance from ashore-based maritime traffic controlstations [5], [6]. However, these navigation aids in the oceanphase have limited facilities; therefore, onboard decision sup-port systems based on intelligent guidance should be developedas proposed in this study.Furthermore, collision avoidance in the ocean phase can

    be divided into two categories: long-range and close-prox-imity collision avoidance. Long-range collision avoidance isfacilitated by the current law of the sea, the Convention onthe International Regulations for Preventing Collisions at Sea(COLREGs) [7], formulated by the IMO. The reported dataon ship collisions show that 56% of major maritime collisionsinvolve violation of the COLREGs rules and regulations [8].However, collision avoidance in close-proximity conditions isnot facilitated by such rules and regulations, but the navigatorsknowledge and experience can play an important role in thosesituations. Therefore, the maintenance of safe distance amongvessels and other obstacles plays the most important role inimproving ship safety in close-proximity conditions. This safedistance keeping among vessels, especially in overtakes andhead-on situations, is also emphasized by the COLREGs [7],[9].However, in close-proximity conditions, vessel-to-vessel

    interaction forces and moments are highly activated, and thatcould also be affected by vessels orientations. These forcesand moments could also result in involuntary course and speedchanges, and that could eventually lead to various ship collisionand near-miss situations. Therefore, these conditions havebeen extensively studied in the recent literature, as furtherdiscussed in this section. A simulation model of two shipspassing under vessel-to-vessel interaction forces and momentson constant parallel course is presented in [10]. However, thesevessel-to-vessel interactions may complicate ship navigationin a narrow channel [11] under vessel traffic [10], [12] andin shallow-water conditions [13], where bank effects, andweather and environmental conditions can also be influential.Therefore, ship behavior under these interaction forces andmoments should be further considered to avoid collision andnear-collision situations under close-proximity conditions inship navigation.The vessel-to-vessel interaction forces in surge and swaymay

    cause vessels to either attract or repulse from each other, andthe yaw moments may cause them to either rotate toward oraway from each other. However, these hydrodynamic forces andmoments could also be affected by several factors: size, lateraland longitudinal separation distance, speeds, wetted hull shapesof the vessels, water depth and transverse distances from thechannel banks when the vessels are close to shore, and weatherconditions [14], [15]. These vessel interaction forces and mo-ments have been calculated by several numerical methods inthe recent literature. Methods for calculation of sway force andyaw moments for two vessels moving under close proximity indeep-water and shallow-water conditions are presented by Tuckand Newman [16] and Yeung [17], respectively. Other numer-ical methods for predicting such forces and moments for two

    vessels moving under close proximity are also presented byHuang and Chen [18], Sutulo and Guedes Soares [19], Xiangand Faltinsen [20], King [21], Varyani and Krishnankutty [22]Xu et al. [23], Sutulo et al. [24], and Zhou et al. [25]. A theo-retical method to predict the sinkage and trim conditions of twomoving vessels under parallel meeting and overtaking condi-tions is presented by Gourlay [26]. Therefore, the possible ac-tions against these vessel interaction forces andmoments shouldbe executed by the navigators as early as possible to avoid closeencounter situations in ship navigation.

    C. Ship Collision Avoidance

    Decision-making processes in ship collision avoidance arepresented in various studies [27][29]. Furthermore, severalstudies have been dedicated to the subject of collision avoid-ance maneuvers based on the following concepts: a clusteredgroup of ships under close-proximity conditions [30]; state,parameter, and action optimization conditions [31][34];safe navigational trajectories/routes selections [35][40];case-based reasoning [41]; intelligent anticollision algorithms[42]; artificial force fields [43], [44]; fuzzy-logic-based systems[45][48]; IFTHEN-logic-based systems [49]; neurofuzzynetworks [50]; and line of sight counteractions [51]. However,one should note that none of the above literature has presentedproper collision avoidance experimental results, which in turnare the main contribution of this study.Therefore, this study proposes intelligent guidance for ship

    collision avoidance in e-navigation environment, which hasbeen initiated in [52][56]. Furthermore, the proposed approachhas been evaluated under an experimental setup, and the resultson several collision avoidance situations of two vessels bymeans of autonomous maneuvers are also presented in thisstudy.The experimental setup consists of a navigation and control

    platform, and a vessel model that is presented in a mathemat-ical formulation as well as in an actual implementation. Thenavigation and control platform consists of controlling the shipmodel in autonomous and manual modes. The vessel model isused to create several collision avoidance situations and thatis supported by an intelligent-guidance-based collision avoid-ance system. The proposed collision avoidance system capa-bilities of making multiple parallel collision avoidance deci-sions regarding several vessel collision situations are also illus-trated. However, those decisions are executed as sequential ac-tions to avoid complex collision situations in ship navigation inlong-distance as well as in close-proximity conditions, which isdiscussed further.

    II. MATHEMATICAL FORMULATION

    A proposed mathematical formulation for ship navigation(i.e., autonomous navigation and collision avoidance) is pre-sented in Fig. 1. It consists of three main systems: the vesseltraffic monitoring and information system (VTMIS), the col-lision avoidance system (CAS), and the vessel control system(VCS).

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    PERERA et al.: EXPERIMENTAL EVALUATIONS ON SHIP AUTONOMOUS NAVIGATION AND COLLISION AVOIDANCE 3

    Fig. 1. Mathematical formulation for ship navigation.

    A. Vessel Traffic Monitoring and Information SystemThe VTMIS facilitates by providing ship traffic information

    (i.e., ships position, course, speed, acceleration, and trajectoryconditions) which can be used for autonomous navigation pur-poses as well as for collision avoidance among ships. Besides ascan sensor (i.e., radar/laser sensor), there are three main mod-ules: vessel detection and tracking (VDT), vessel state estima-tion and trajectory prediction (VSETP), and intervessel commu-nication (IVC).A scan sensor is used for detecting vessel positions. An ar-

    tificial neural network (ANN)-based multivessel detection and

    tracking process has been implemented on the VDT module. Itdetects and tracks ships navigating in the scan sensor vicinity.An extended Kalman filter (EKF)-based vessel state estimation(i.e., position, velocity, and acceleration) and navigational tra-jectory prediction process has been implemented on the VSETPmodule. This process is executed under the information given bythe VDT module. The vessel traffic information (i.e., ship posi-tion, course, speed, etc.) transfers among ships and shore-basedmaritime authorities and that could be managed by the IVCmodule through a wireless network. An extensive study on theVTMIS considered in this paper is presented in [6].

    B. Collision Avoidance SystemThe CAS generates collision avoidance decisions/actions in

    a sequential format that can be executed in ship navigation. Itis expected to have this system installed onboard a vessel thatis called as the own vessel for the autonomous navigationand collision avoidance experiments. As presented in Fig. 1,the CAS consists of four modules: own-vessel communication(OVC), parallel decision making (PDM), sequential action for-mation (SAF), and collision risk assessment (CRA).The OVC module facilitates the communication of naviga-

    tion information among ships and VTMISs. Such data are usedby the PDMmodule to make collision avoidance decisions. ThePDM module consists of a fuzzy-logic-based decision-makingprocess that generates parallel collision avoidance decisionswith respect to each ship that is under collision course with theown vessel. Furthermore, that creates course and speed changedecisions for the own vessel, upon which decisions transfer tothe SAF module to create proper collision avoidance actions.The rules, regulations, and expert navigational knowledgeproposed by the COLREGs have been considered in the imple-mentation of the PDM module. An extensive discussion on thismodule is presented in [54] and [56].The CRA module evaluates the collision risk and the ex-

    pected time until collision of each ship with respect to the ownvessel based on navigation information from the OVC module.The evaluated collision risk information is transferred to theSAF module to arrange collision avoidance actions appropri-ately. An extensive discussion of the CRA module is presentedin [57] and [58].The SAF module converts the parallel collision avoidance

    decisions that are initially generated by the PDM module intosequential actions, considering the time until collision for eachcollision situation estimated by the CRA module. An extensivediscussion on the SAF module is presented in [55] and [56].Finally, the sequential collision avoidance actions that are orga-nized by the SAF module are shared with the VCS.These actions can be categorized into two sets, course and

    speed controls, that will be implemented on the own vessel.The course and speed control collision avoidance actions withrespect to each collision situation are executed under two sub-systems: the steering control subsystem (SCS) and the speedcontrol subsystem (SPS). The SCS and the SPS control shipcourse and speed conditions, respectively. An overview of thecollision avoidance decision/action execution process under thePDM and SAFmodules is discussed further in Sections II-C andII-D.

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    4 IEEE JOURNAL OF OCEANIC ENGINEERING

    Fig. 2. PDM module.

    C. Parallel Decision Making Module

    The PDM module consists of three main units (see Fig. 2):fuzzification, fuzzy rules, and defuzzification. The inputs of theOVCmodule, namely, range, bearing, course, and speed of othervessels (i.e., target vessels) for which there are collision courseswith the own vessel at their respective instants, are fuzzified inthe fuzzification unit.Accordingly, the following input fuzzy membership func-

    tions (FMFs) are considered: range FMF, speed ratio FMF,bearing FMF, and relative course FMF. Afterwards, the fuzzi-fied results are transferred into the fuzzy rules unit for furtheranalysis. Mamdani type IFTHEN rules are developed andinference via MinMax norm is considered in the fuzzy rulesunit. As mentioned before, the IFTHEN fuzzy rules are devel-oped in accordance with the COLREGs rules and regulations.However, expert navigational knowledge is also considered inthe fuzzy rules development process.The course and speed change decisions to avoid the target

    vessels that have collision course with the own vessel aregenerated by the defuzzification unit. The inference resultsfrom the fuzzy rules unit are defuzzified by considering thefollowing output FMFs: course change FMF and speed changeFMF. These FMFs generate course and speed change decisionsthat will be executed for collision avoidance in the own vessel.An extensive discussion on fuzzification, fuzzy rules, anddefuzzification related to the present approach is presented in[54] and [56].

    Fig. 3. SAF module.

    D. Sequential Action Formulation Module

    The SAF module that is modeled as a Bayesian network con-sists of four nodes/units (see Fig. 3): time until collision estima-tion (TUCE), collision risk estimation (CRE), collision avoid-ance action formulation (CAAF), and action delay. The mainobjective of the SAF module is to transform the parallel colli-sion avoidance decisions that are generated by the PDMmoduleinto sequential actions that should be executed in the own vessel.This can be achieved by collecting the collision avoidance de-cisions and evaluating them using the time until collision withrespect to each vessel that has collision course with the ownvessel. Then, the final results (i.e., collision avoidance actions)are arranged as a sequential formation involving the course andspeed change actions at the respective instants.The inputs of the SAF module are the collision decisions and

    the collision risk generated by the PDM and CRA modules, re-spectively. Themain objectives of the TUCE andCRE nodes areto estimate the time until collision and the collision risk betweenthe own and target vessels, respectively. The actions delay is de-signed to formulate the appropriate time interval for executingthe speed and course change actions to avoid each collision sit-uation.Therefore, the vessel collision avoidance actions are formu-

    lated by the CAAF node, and that is affected by the action delayand the CRE nodes, as presented in Fig. 3. Such actions canbe divided into two sections: course and speed change actionswhich are initially generated as the collision avoidance deci-sions from the PDM module. Finally, these accumulated ac-tions are implemented in the VCS of the own vessel for col-lision avoidance among vessels. An extensive discussion on theBayesian-network-based sequential collision avoidance actionformulation is presented in [55] and [56]. The decisions/actionsthat need to be taken by the own vessel to avoid various colli-sion situations are summarized in Table I.

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    PERERA et al.: EXPERIMENTAL EVALUATIONS ON SHIP AUTONOMOUS NAVIGATION AND COLLISION AVOIDANCE 5

    TABLE ICOLLISION AVOIDANCE DECISIONS/ACTIONS

    Fig. 4. Ship model in autonomous navigation and collision avoidance maneu-vers.

    III. EXPERIMENTAL SETUPThe experimental setup, which is further discussed in this

    section, consists of a navigation and control platform, and thevessel model. This model consists of a scaled ship that hasbeen used in this study; and the navigation and control plat-form has been used for autonomous and manual control of thevessel model. Therefore, the setup has been used to conduct shipmaneuvers in autonomous navigation and collision avoidanceexperiments. The proposed CAS is implemented on the vesselmodel, which is considered as the own vessel.

    A. Vessel ModelThe vessel model considered in this study is presented in

    Fig. 4 and its characteristics are as follows: overall length of2.590 m; length between perpendiculars of 2.450 m; breadthequal to 0.430 m; depth of 0.198 m; and estimated trail draftand displacement of 0.145 m and 115.6 kg, respectively. Thevessel model is built in single skin glass reinforced polyesterwith plywood framings and that is controlled by the navigationand control platform, which can be divided into two sections:hardware structure and software architecture.

    Fig. 5. Command and monitoring unit.

    B. Hardware Structure

    The hardware structure consists of all sensors and actuatorsthat are used in the navigation and control platform. This struc-ture is further divided into the following two units: the commandand monitoring unit (CMU) and the communication and con-trol unit (CCU). The main objective of the CMU is to facilitatemanual and autonomous control of the vessel model providedby the humanmachine interface (HMI), as presented in Fig. 5.The CCU is implemented on a shore-based station and that con-sists of several instrumentations such as a laptop computer, aGlobal Positioning System (GPS) unit, and an industrial WiFiunit.A laptop computer, used as HMI, is connected to an indus-

    trial WiFi unit for communicating with the CCU. The computerworks as a data display interface as well as an automaticand manual control unit for the vessel model. Furthermore,the above discussed VTMIS is implemented on the laptopcomputer under MATLAB/LABVIEW software. The VTMISis simulated to obtain the target vessel behavior that is in acollision course with the vessel model (i.e., own vessel). Thedata are forwarded to the CAS for collision avoidance deci-sions/actions. One should note that the CAS is implemented onthe vessel model (i.e., own vessel).The GPS unit is used in the CMU for position measurements

    of the vessel model. The complete GPS system has two units,namely, a base station and a rover station which improve the po-sition accuracy of the vessel model. The base GPS station unitacts as a stationary reference that transmits known stationaryposition correction signals for the rover GPS station which islocated in the ship model. The WiFi unit is used for communi-cation between the ashore-based CMU and the onboard CCU.The proposed CCU is implemented on the vessel model as

    presented in Fig. 6. The main objective of the CCU is to executethe fuzzy-Bayesian-based decision/action execution process(i.e., the CAS), as described in Section II. That is associatedwith the course and speed change actions and is facilitatedby the following instrumentation: two CompacRIO units, anindustrial Ethernet switch (IES), a laptop computer, a GPS unit,an inertial measurement system (IMS), a WiFi unit, and twodirect current (dc) motors.Two CompactRIOs with input/output (I/O) modules are used

    in the CCU. One collects digital data from the IMS and GPSunits. Other unit is connected to the steering and speed controlsubsystems of the vessel model to control the actuations of therudder and propeller assembled to two dc motors. Furthermore,both CompactRIOs are connected through the IES.

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    6 IEEE JOURNAL OF OCEANIC ENGINEERING

    Fig. 6. Communication and control unit.

    Fig. 7. Trajectories of collision situation I.

    The laptop computer in the vessel model is used to recordand store the digital data collected from the IMS and GPS unitsthrough the IES connected to the CompactRIO. Another on-board GPS unit is used in the CCU to accurately estimate theposition of the vessel model as discussed previously. The IMS

    unit consists of the following sensors: a magnetometer, an ac-celerometer, rate gyro, and a GPS receiver. The IMS is capableof measuring the following: three-axes angles of heading, roll,and pitch; three-axes angular velocities of heading, roll, andpitch; and three-axes linear accelerations of surge, sway, andheave. The internal GPS receiver in the IMS unit measures thevessel model position facilitated with WAAS capabilities. TheIES is used in the CCU as a communication gateway amongsensors, actuators, and CompactRIO units.Furthermore, the above discussed CAS is implemented on the

    laptop under MATLAB/LABVIEW software. The CAS formu-lates the collision avoidance actions that are based on the targetvessel collision course and speed information that is given bythe shore-based VTMIS (i.e., other laptop computer). AnotherWiFi unit is used for communication between the ashore-basedCMU and the onboard CCU connected through the IES.The proposed vessel model has two control subsystems in-

    corporated: the steering control subsystem (SCS) and the speedcontrol subsystem (SPS). The SCS is associated to the ruddercontrol system, and its main objective is to maintain the appro-priate vessel course during its maneuvers. The SPS is associ-ated to the propeller control system, and its main objective isto maintain appropriate vessel speed during its maneuvers. Theproportionalintegralderivative (PID) controllers are used forboth propeller revolutions per minute (RPM) and rudder posi-tions controls. The course and speed change collision avoidanceactions that are generated by the CAS are executed in these sub-systems.

    C. Software ArchitectureThe software architecture in this experimental platform is

    mainly developed under LABVIEW and MATLAB programsconsisting of several loops: a field-programmable gate array(FPGA) loop, a real-time control loop, a CAS loop, and aTCP/IP loop. The FPGA loop aims at collecting data fromthe sensors (i.e., GPS and IMS units) and controlling the ac-tuations of the steering and speed subsystems that have beenprogrammed under LABVIEW.The associated PID controllers for the steering and speed con-

    trol subsystems are implemented under a real-time control loop(i.e., the internal deterministic control loop) that has the highestresponsiveness, determinism, and priority with comparison toother software loops. The data processing and record saving forthe respective sensors are implemented under the internal non-deterministic loop that has lower priority in comparison to thedeterministic control loop.The CAS loop consists of the proposed fuzzy-Bayesian-based

    decision/action execution process for collision avoidanceamong vessels, generating required collision avoidance actionsfor the vessel model with respect to the simulated target vessel,with which it is in a collision course. These actions are executedunder the real-time control loop associated with the steeringand speed control subsystems on the vessel model. The TCP/IPloop is related to the communication between shore-basedCMU and the VTMIS, being implemented under wireless com-munication through the industrial WiFi unit. Furthermore, anextensive discussion on the experimental platform is presentedin [59] and [60].

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    PERERA et al.: EXPERIMENTAL EVALUATIONS ON SHIP AUTONOMOUS NAVIGATION AND COLLISION AVOIDANCE 7

    Fig. 8. Vessels positions, speed, and course control of collision situation I.

    IV. EXPERIMENTAL RESULTS

    The collision avoidance experiments were conducted on thelake of Campo Grande in Lisbon, Portugal. These experi-ments involved various autonomous maneuvers and collisionavoidance situations by the vessel model. The collision avoid-ance experiments were conducted on the ship model with theonboard CAS, as described in Section II. However, a scaled ver-sion of the CAS has been used during these experiments due tothe practical difficulties (i.e., wind and wave conditions) facedby the vessel model. Furthermore, the vessel model positiondata were collected from the GPS system that has two units (i.e.,a base station and a rover station) due to its higher accuracy (i.e.,1 cm). However, the additional sensor data (i.e., IMS sensor)have encountered lower accuracy due to the sensor noise andslow speed conditions of the vessel model.The vessel model with the CAS was represented as the own

    vessel, and a target vessel in a collision course with the ownvessel was simulated. However, it was observed that the formu-lation of a collision situation between two ships is extremelydifficult to achieve due to the ship model sudden course changeand speed variations caused by the wind and wave conditions.Therefore, an additional algorithm was developed to simulatethe target vessel maneuvers.This target vessel algorithm consists of the following sequen-

    tial steps: the initial target vessel position should be assignednear the own vessel navigation route; then, the algorithm createsproper collision course between the own and target vessels by

    Fig. 9. Trajectories of collision situation II.

    considering various target vessel course conditions. As an ex-ample, the target vessel course changes from 0 to 360 with 1intervals under the speed condition that is approximated to theown vessel speed; when the algorithm finds a collision course

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    8 IEEE JOURNAL OF OCEANIC ENGINEERING

    Fig. 10. Vessels positions, speed, and course control of collision situation II.

    between the two vessels during such course changes, it executesthat course with the appropriate speed conditions as the targetvessel. Consequently, this target vessel algorithm generates anappropriate collision situation between both vessels, and that in-formation is forwarded to the CAS. Therefore, the own vesseltakes appropriate decisions/actions to avoid the collision situa-tion.Several collision situations were created by the proposed

    target vessel algorithm, and the appropriate actions taken bythe vessel model were observed under such conditions. TheCAS was implemented on the laptop computer onboard theown vessel, as described in Section II. It is assumed that thetarget vessel is moving at constant speed and course conditionsand does not honor any navigational rules and regulations(i.e., COLREGs). One should note that such speed and courseconditions are considered in these experiments to keep theconsistence in the collision situation between two vessels [56].

    A. Collision Situation I

    The first set of experimental results of a collision situation be-tween two vessels is presented in Figs. 7 and 8. As presented inFig. 7, the vessel model (i.e., own vessel) and the target vesselstart to navigate from the positions (0 [m], 0 [m]) and (10 [m],20 [m]), respectively. One should note that the spiral section ofthe target vessel trajectory, which is near its initial position, rep-

    Fig. 11. Trajectories of collision situation III.

    resents the algorithm that has been used to capture the collisioncourse between two vessels, as described previously.

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    PERERA et al.: EXPERIMENTAL EVALUATIONS ON SHIP AUTONOMOUS NAVIGATION AND COLLISION AVOIDANCE 9

    Fig. 12. Vessels positions, speed, and course control of collision situation III.

    As presented in Fig. 7, the vessel model (i.e., the own vessel)has observed a possible collision situation in which the targetvessel is approaching for a crossing situation from starboard.One should note that in accordance with the COLREGs rulesand regulations the vessel model is in a give way situation, inwhich it has lower priority for navigation in a collision situation,and the target vessel is in a stand on situation. Therefore, theearly collision avoidance actions to avoid the collision situationsare executed by the vessel model. The respective own and targetvessel coordinates with respect to time are presented inthe top plots of Fig. 8. The collision avoidance decisions (seeTable I) of altering course to starboard and increasing speed atthe first stage, altering course to port and increasing speed at thesecond stage, altering course to starboard and increasing speedat the third stage, which have been taken by the ship model, arepresented in the bottom plots of Fig. 8.

    B. Collision Situation IIThe second set of experimental results of a collision situation

    with two vessels is presented in Figs. 9 and 10. The vessel modeland the target vessel start to navigate from the positions (0 [m],0 [m]) and (20 [m], 10 [m]), respectively. As presented in Fig. 9,the vessel model (i.e., the own vessel) has observed a possiblecollision situation in which the target vessel is approaching fora crossing situation from port. With respect to the COLREGsrules and regulations, the vessel model is in a stand on situ-ation, thus it has higher priority for navigating, and the target

    Fig. 13. Trajectories of collision situation IV.

    vessel is in a give way situation, meaning a lower priority fornavigating in a collision situation. Since the target vessel has thegive way situation and the vessel does not take any actions to

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    10 IEEE JOURNAL OF OCEANIC ENGINEERING

    Fig. 14. Vessels positions, speed, and course control of collision situation IV.

    avoid the collision situation, the vessel model is forced to takeappropriate actions in that sense.It must be noted that the own vessel in a give way situa-

    tion takes collision avoidance actions earlier than in a standon situation. In this context, the distance between two ves-sels has been considered for the decision-making process inthe CAS. Therefore, the vessel in a stand on situation couldmake crash-stop type maneuvers to avoid a collision situationdue to inadequate actions from the target vessel. These situa-tions have been categorized as critical collision conditions, andabove discussed concepts have been adopted by the CAS, asfurther described in [48]. The respective own and target vessel coordinates with respect to time are presented in the topplots of Fig. 10. The collision avoidance decisions (see Table I)of altering course to starboard and reducing speed in the firststage and altering course to starboard and increasing speed atthe second stage that have been taken by the vessel model arepresented in the bottom plots of Fig. 10.

    C. Collision Situation IIIThe third set of experimental results of a collision situation

    with two vessels is presented in Figs. 11 and 12. The vesselmodel and the target vessel start to navigate from the positions(0 [m], 0 [m]) and ( 10 [m], 20 [m]), respectively. As pre-sented in Fig. 11, the vessel model (i.e., the own vessel) has ob-served a possible collision situation in which the target vesselis approaching for a head-on collision situation from starboard.

    Fig. 15. Trajectories of collision situation V.

    However, the target vessel does not take any action in this closeencounter situation; therefore, the vessel model is forced to takeappropriate actions to avoid the collision situation. Even though

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    PERERA et al.: EXPERIMENTAL EVALUATIONS ON SHIP AUTONOMOUS NAVIGATION AND COLLISION AVOIDANCE 11

    Fig. 16. Vessels positions, speed, and course control of collision situation V.

    the vessels should pass port to port in a head-on collision situ-ation in accordance with the COLREGs [7], [9], this close en-counter situation with an altering course to starboard by the ownvessel could increase the collision risk. Therefore, this safe dis-tance keeping among vessels, especially in close encounter sit-uations, is emphasized by the COLREGs and is implementedby the vessel model in this situation. The respective own andtarget vessel coordinates with respect to time are presentedin the top plots of Fig. 12. The collision avoidance decisions (seeTable I) of altering course to port and increasing speed taken bythe own vessel are presented in the bottom plots of Fig. 12.

    D. Collision Situation IV

    The fourth set of experimental results of a collision situationwith two vessels is presented in Figs. 13 and 14. The vesselmodel and the target vessel start to navigate from the positions(0 [m], 0 [m]) and (10 [m], 20 [m]), respectively. As presentedin Fig. 13, the vessel model (i.e., the own vessel) has observeda possible collision situation in which the target vessel is ap-proaching for a crossing situation from starboard. According tothe COLREGs rules and regulations, the own and target vesselsare in give way and stand on situations, respectively. There-fore, the early collision avoidance actions to avoid the collisionsituations are executed by the vessel model. The respective ownand target vessel coordinates with respect to time are pre-sented in the top plots of Fig. 14.

    The collision avoidance decisions (see Table I) of alteringcourse to starboard and increasing speed at the first stage, al-tering course to port and increasing speed at the second stage,and altering course to starboard and increasing speed at the thirdstage, which have been taken by the own vessel, are presented inthe bottom plots of Fig. 14. Considerable similarities can also benoted on the collision avoidance actions executed by the vesselmodel in the collision situations I and IV.

    E. Collision Situation VThe fifth set of experimental results of a collision situation

    with two vessels is presented in Figs. 15 and 16. The vesselmodel and the target vessel start to navigate from the positions(0 [m], 0 [m]) and (30 [m], 10 [m]), respectively. As presented inFig. 15, the vessel model (i.e., own vessel) has observed a pos-sible collision situation in which the target vessel is approachingfor a head-on collision situation from port.However, the target vessel does not take any action in this

    close encounter situation, therefore the ship model is forced totake appropriate actions to avoid the collision situation. Eventhough the vessels should pass in port to port in a head-on col-lision situation in accordance with the COLREGs [7], [9], thisclose encounter situation with altering course to starboard bythe own vessel could also increase the collision risk. Therefore,the safe distance keeping among vessels, especially in closeencounter situations, is emphasized by the COLREGs and isalso considered in this situation. The respective own and target

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    12 IEEE JOURNAL OF OCEANIC ENGINEERING

    TABLE IIAPPROACHES FOR POSSIBLE COLLISIONS AND RESPECTIVE COLREGS RULES AND REGULATIONS

    TABLE IIISHIP MODEL COURSE AND SPEED COLLISION AVOIDANCE DECISIONS

    vessel coordinates with respect to time are presented inthe top plots of Fig. 16. The collision avoidance decisions (seeTable I) of altering course to port and increasing speed at thefirst stage and altering course to starboard and increasing speedat the second stage, which have been taken by the ship model,are presented in the bottom plots of Fig. 16.In the experimental results, one can observe five possible col-

    lision situations that the vessel model (i.e., the own vessel) hasidentified. This is resumed in Table II along with the properCOLREGs rules and regulations for the vessel model and thetarget vessel. It is assumed that the target vessel does not honorany navigational rules and regulations (i.e., COLREGs). Eventhough the purpose of these experiments is for the vessel modelto always take appropriate collision avoidance actions, the factis that in some situations the own vessel may not be the onewho has the priority to take collision avoidance actions in thefirst place, according to the COLREGs rules and regulations.These collision avoidance actions with respect to the COLREGsrules and regulations have been summarized in Table II, and theaction stages that have been executed by the vessel model aresummarized in Table III for each situation.

    V. CONCLUSION AND FUTURE DEVELOPMENT

    Experimental evaluations on several collision avoidance sit-uations between two vessels have been presented in this study.Considering the experimental results, it can be concludedthat the ship model has taken appropriate collision avoidancedecisions and actions to reduce the collision risk betweenboth vessels. Therefore, the reported successful experimentalresults using a vessel model show the superior capabilities ofthe proposed intelligent-guidance-based collision avoidancesystem, and this is the main contribution of this study.Furthermore, the tools and techniques presented in the study

    can be used for the e-navigation strategy, in which one couldintroduce autonomous navigation and collision avoidance func-tionalities in the shipping industry. However, the implementa-tion of collision detection and avoidance among multiple ves-sels in the experimental platform is still a challenge for the fu-ture, where the proposed system should be further developed.Therefore, a complete version of the proposed collision detec-tion and avoidance under multivessel situations will be a furtherdevelopment of this study.

    REFERENCES

    [1] N. Ward and S. Leighton, Collision avoidance in the e-navigation en-vironment, in Proc. 17th Conf. Int. Assoc. Marine Aids Navig. Light-house Authorities, Cape Town, South Africa, 2010, pp. 410.

    [2] IMO, Development of an e-navigation strategy, Report of the corre-spondence group on e-navigation, NAV/53/13, 2007.

    [3] MUNIN, Maritime unmanned navigation through intelligence innetworks, 2013 [Online]. Available: http://www.unmanned-ship.org/munin/

    [4] R. S. Burns, G. Blackwell, and S. Calvert, An automatic guidance,navigation and collision avoidance system for ships at sea, in Proc.IEE Colloq. Control Mar. Ind., Jan. 1988, pp. 3/13/3.

    [5] A. Janex, Concept of a collision-avoidance system for marine naviga-tion, in Proc. OCEANS Conf., Sep. 1990, pp. 458463.

    [6] L. Perera, P. Oliveira, and C. Guedes Soares, Vessel detection,tracking, state estimation and navigation trajectory prediction for thevessel traffic monitoring and information process, IEEE Trans. Intell.Transp. Syst., vol. 13, no. 3, pp. 11881200, 2012.

    [7] IMO, Convention on the International Regulations for PreventingCollisions at Sea (COLREGs), 1972.

    [8] T. Statheros, G. Howells, and K. McDonald-Maier, Autonomous shipcollision avoidance navigation concepts, technologies and techniques,J. Navig., vol. 61, pp. 129142, 2008.

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    PERERA et al.: EXPERIMENTAL EVALUATIONS ON SHIP AUTONOMOUS NAVIGATION AND COLLISION AVOIDANCE 13

    [9] L. Perera, J. P. Carvalho, and C. Guedes Soares, Autonomous guid-ance and navigation based on the COLREGs rules and regulations ofcollision avoidance, in Advanced Ship Design for Pollution Preven-tion, C. Guedes Soares and J. Parunov, Eds. London, U.K.: Taylor &Francis, 2010, pp. 129142.

    [10] J. C. Hewlett, Ship Navigation Simulation Study, Houston-GalvestonNavigation Channels, U.S. Army Engineer District, Galveston, TX,USA, AD-A279 016, 1994.

    [11] Y. Yavin, C. Frangos, G. Zilman, and T. Miloh, Computation of fea-sible command strategies for the navigation of a ship in a narrow zigzagchannel, Comput. Math. Appl., vol. 30, no. 10, pp. 79101, 1995.

    [12] K. S. Varyani, A. Thavalingam, and P. Krishnankutty, New genericmathematical model to predict hydrodynamic interaction effects forovertaking maneuvers in simulations, J. Mar. Sci. Technol., vol. 9,pp. 2431, 2004.

    [13] X. Zhou, S. Sutulo, and C. Guedes Soares, Computation of ship-to-ship interaction forces by a 3d potential flow panel method in finitewater depth, in Proc. 29th Int. Conf. Ocean Offshore Arctic Eng.,Shanghai, China, 2010, OMAE2010-20497.

    [14] S. Sutulo and C. Guedes Soares, A unified nonlinear mathematicalmodel for simulating ship manoeuvring and seakeeping in regularwaves, in Proc. Int. Conf. Mar. Simul. Ship Manoeuv., Terschelling,The Netherlands, 2006.

    [15] R. Skejic and M. Faltinsen, A unified seakeeping and maneuveringanalysis of two interacting ships, in Proc. 2nd Int. Conf. Mar. Res.Transp., Naples, Italy, Jun. 2007, pp. 209218.

    [16] E. O. Tuck and J. N. Newman, Hydrodynamic interaction betweenships, in Proc. 10th Symp. Naval Hydrodyn., Cambridge, MA, USA,1974, pp. 3558.

    [17] R. W. Yeung, On the interaction of slender ships in shallow water, J.Fluid Mech., vol. 85, pp. 143159, 1978.

    [18] E. T. Huang and H. C. Chen, Passing ship effects on moored vessels atpiers, in Proc. First Symp. CSLC Prevention, Long Beach, CA, USA,2006.

    [19] S. Sutulo and C. Guedes Soares, Simulation of the hydrodynamicinteraction forces in close-proximity manoeuvring, in Proc. 27thInt. Conf. Offshore Mech. Arctic Eng., Estoril, Portugal, 2008,OMAE2008-57938.

    [20] X. Xiang and O. M. Faltinsen, Maneuvering of two interacting shipsin calm water, in Proc. 11th Int. Symp. Practical Design Ships OtherFloating Struct., Rio da Janeiro, Brazil, 2010, pp. 161171.

    [21] G. W. King, Unsteady hydrodynamic interactions between ships inshallow water, in Proc. 6th Austral. Hydraul Fluid Mech. Conf., Ade-laide, Australia, 1977, pp. 291293.

    [22] K. S. Varyani and P. Krishnankutty, Modification of ship hydrody-namic interaction forces and moment by underwater ship geometry,Ocean Eng., vol. 33, pp. 10901104, 2006.

    [23] Y. Xu, Z. Zou, M. Liu, and K. S. Varyani, Study on critical uncon-trollable hydrodynamic interaction between ships, in Proc. 18th Int.Offshore Polar Eng. Conf., Vancouver, BC, Canada, 2008.

    [24] S. Sutulo, C. Guedes Soares, and J. F. Otzen, Validation of poten-tial-flow estimation of interaction forces acting upon ship hulls inside-to-side motion, J. Ship Res., vol. 56, no. 3, pp. 129145, 2012.

    [25] X. Zhou, S. Sutulo, and C. Guedes Soares, Computation of ship hydro-dynamic interaction forces in restricted waters using potential theory,J. Mar. Sci. Appl., vol. 11, pp. 265275, 2012.

    [26] T. Gouraly, Sinkage and trim of two ships passing each other on par-allel course, Ocean Eng., vol. 36, no. 14, pp. 11191127, 2009.

    [27] C. Chauvin and S. Lardjane, Decision making and strategies in aninteraction situation: Collision avoidance at sea, Transp. Res. F, vol.11, no. 4, pp. 259262, 2008.

    [28] Y. Y.Wang, A. K. Debnath, and H. C. Chin, Modeling collision avoid-ance decisions in navigation, in Proc. 10th Asian Conf. Mar. Simul.Simul. Res., Keelung, Taiwan, Jun. 2010, pp. 191198.

    [29] M. R. Benjamin and J. A. Curcio, COLREGS-based navigation of au-tonomous marine vehicles, in Proc. IEEE/OES Autonom. UnderwaterVeh., Jun. 2004, pp. 3239.

    [30] A. Kawaguchi, X. Xiong, M. Inaishi, and H. Kondo, A computer-ized navigation support for maneuvering clustered ship groups in closeproximity, Syst. Cybern. Inf., vol. 3, no. 3, pp. 4656, 2006.

    [31] A. Miele, T. Wang, C. S. Chao, and J. B. Dabney, Optimal control ofa ship for collision avoidance maneuvers, J. Optim. Theory Appl., vol.103, no. 3, pp. 495519, Dec. 1999.

    [32] T. Miloh and S. D. Sharma, Critical maneuvers for avoiding collisionat sea, Inst. Schiffbau der Univ. Hamburg, Hamburg, Germany, Apr.1975.

    [33] S. D. Sharma, On ship maneuverability and collision avoidance, inProc. 2nd West Eur. Mar. Technol. Conf., London, U.K., May 1977,pp. 129.

    [34] T. L. Vincent, Collision avoidance at sea, in Differential Games andApplications, ser. Lecture Notes in Control and Information Sciences,P. Hagedorn, Ed. Berlin, Germany: Springer-Verlag, 1977, vol. 3, pp.205221.

    [35] R. Smierzchalski and Z. Michalewicz, Modeling of ship trajectory incollision situations by an evolutionary algorithm, IEEE Trans. Evol.Comput., vol. 4, no. 3, pp. 227241, Sep. 2000.

    [36] M. Ito, F. Zhang, and N. Yoshida, Collision avoidance control of shipwith genetic algorithm, in Proc. IEEE Int. Conf. Control Appl., Aug.1999, pp. 17911796.

    [37] J. Froese and S. Mathes, Computer-assisted collision avoidance usingARPA and ECDIS, German J. Hydrogr., vol. 49, no. 4, pp. 519529,1997.

    [38] X. Zeng,M. Ito, and E. Shimizu, Building an automatic control systemof maneuvering ship in collision situation with genetic algorithms, inProc. Amer. Control Conf., Jun. 2001, pp. 28522853.

    [39] X. Hong, C. J. Harris, and P. A. Wilson, Autonomous ship colli-sion free trajectory navigation and control algorithms, in Proc. 7thIEEE Int. Conf. Emerging Technol. Factory Autom., 1999, vol. 2, pp.923929.

    [40] X. D. Cheng, Z. Y. Liu, and X. T. Zhang, Trajectory optimizationfor ship collision avoidance system using genetic algorithm, in Proc.OCEANS Conf. Asia Pacific, May 2006, DOI: 10.1109/OCEANSAP.2006.4393976.

    [41] Y. Liu and H. Liu, Case learning base on evaluation system for vesselcollision avoidance, in Proc. 5th Int. Conf. Mach. Learn. Cybern.,Aug. 2006, pp. 20642069.

    [42] S. Yang, L. Li, Y. Suo, and G. Chen, Study on construction of simula-tion platform for vessel automatic anti-collision and its test methods,in Proc. IEEE Int. Conf. Autom. Logistics, Aug. 2007, pp. 24142419.

    [43] S. Lee, K. Kwon, and J. Joh, A fuzzy logic for autonomous naviga-tion of marine vehicles satisfying COLREG guidelines, Int. J. ControlAutom. Syst., vol. 2, no. 2, pp. 171181, Jun. 2004.

    [44] Y. Xue, B. S. Lee, and D. Han, Automatic collision avoidance ofships, J. Eng. Maritime Environ., vol. 223, no. 1, pp. 3346, 2009.

    [45] Y. I. Lee andY. G. Kim, A collision avoidance system for autonomousship using fuzzy relational products and COLREGs, in Proc. 5th Int.Conf. Intell. Data Eng. Autom. Learn., Exeter, U.K., Aug. 2527, 2004,pp. 247252.

    [46] K. Hasegawa, Automatic collision avoidance system for ship usingfuzzy control, in Proc. 8th Ship Control Syst. Symp., 1987, pp.234258.

    [47] Y. Zhuo and G. E. Hearn, A ship based intelligent anti-collision deci-sion-making support system utilizing trial manoeuvers, in Proc. Chin.Control Decision Conf., Jul. 2008, pp. 39823987.

    [48] L. P. Perera, J. P. Carvalho, and C. Guedes Soares, Fuzzy-logic baseddecision making system for collision avoidance of ocean navigationunder critical collision conditions, J. Mar. Sci. Technol., vol. 16, pp.8499, 2011.

    [49] G. P. Smeaton and F. P. Coenen, Developing an intelligent marinenavigation system, Comput. Control Eng. J., vol. 1, no. 2, pp. 95103,Mar. 1990.

    [50] P. A. Wilson, C. J. Harris, and X. Hong, A line of sight counteractionnavigation algorithm for ship encounter collision avoidance, J. Navig.,vol. 56, no. 1, pp. 111121, Jan. 2003.

    [51] C. J. Harris, X. Hong, and P. A. Wilson, An intelligent guidance andcontrol system for ship obstacle avoidance, Proc. IMechE I/J. Syst.Control Eng., vol. 213, no. 14, pp. 311320, 1999.

    [52] L. P. Perera, J. P. Carvalho, and C. Guedes Soares, Solu-tions to the failures and limitations of Mamdani fuzzy infer-ence in ship navigation, IEEE Trans. Veh. Technol., 2013, DOI:10.1109/TVT.2013.228830.

    [53] L. P. Perera, J. P. Carvalho, and C. Guedes Soares, Mamdani typefuzzy inference failures in navigation, in Proc. 9th IEEE Int. Conf.Ind. Inf., Lisbon, Portugal, Jul. 2011, pp. 2629.

    [54] L. P. Perera, J. P. Carvalho, and C. Guedes Soares, Fuzzy-logic basedparallel collisions avoidance decision formulation for an ocean navi-gational system, in Proc. 8th IFAC Conf. Control Appl. Mar. Syst.,Rostock, Germany, Sep. 2010, pp. 295300.

    [55] L. P. Perera, J. P. Carvalho, and C. Guedes Soares, Bayesian networkbased sequential collision avoidance action execution for an ocean nav-igational system, in Proc. 8th IFAC Conf. Control Appl. Mar. Syst.,Rostock, Germany, Sep. 2010, pp. 301306.

  • This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

    14 IEEE JOURNAL OF OCEANIC ENGINEERING

    [56] L. P. Perera, J. P. Carvalho, and C. Guedes Soares, Intelligent oceannavigation & fuzzy-Bayesian decision-action formulation, IEEE J.Ocean. Eng., vol. 37, no. 2, pp. 204219, Apr. 2012.

    [57] L. P. Perera and C. Guedes Soares, Vector-product based collisionestimation and detection in e-navigation, in Proc. 9th IFAC Conf. Ma-noeuv. Control Marine Craft, Arenzano, Italy, Sep. 2012.

    [58] L. P. Perera and C. Guedes Soares, Detections of potential collisionsituations by relative motions of vessels under parameter uncertain-ties, in Sustainable Maritime Transportation and Exploitation of SeaResources, E. Rizzuto and C. Guedes Soares, Eds. London, U.K.:Taylor & Francis, 2012, pp. 705713.

    [59] L. P. Perera, L. Moreira, F. P. Santos, V. Ferrari, S. Sutulo, and C.Guedes Soares, A navigation and control platform for real-time ma-noeuvring of autonomous ship models, in Proc. 9th IFAC Conf. Ma-noeuv. Control Mar. Craft, Arenzano, Italy, 2012.

    [60] L. P. Perera, V. Ferrari, F. P. Santos, M. A. Hinostroza, and C. GuedesSoares, Experimental results on collisions avoidance of autonomousshipmanoeuvres, inProc. 32nd Int. Conf. OceanOffshore Arctic Eng.,Nantes, France, Jun. 2013, OMAE2013-11265.

    Lokukaluge Prasad Perera received the B.Sc. andM.Sc. degrees in mechanical engineering from Okla-homa State University, Stillwater, OK, USA, in 1999and 2001, respectively, and the Ph.D. degree in navalarchitecture and marine engineering from the Tech-nical University of Lisbon, Lisbon, Portugal, in 2012.He has won Doctoral and Postdoctoral Fellow-

    ships from the Portuguese Foundation for Scienceand Technology in 2008 and 2012 respectively.Currently, he is a Development Engineer at WrtsilFinland Oy, Turku, Finland. His research interests are

    in maritime systems, instrumentation, guidance and control, condition-basedmonitoring, energy efficiency, and emission control, safety, risk, and reliability.

    Victor Ferrari received the B.Sc. and M.Sc. degreesin naval architecture and marine engineering fromthe University of Genoa, Genova, Italy, in 2006 and2009, respectively. He is currently working towardthe Ph.D. degree in naval architecture and marineengineering at the Centre for Marine Technology andEngineering (CENTEC), Instituto Superior Tcnico,Technical University of Lisbon, Lisbon, PortugalIn 2011 and 2012, he was a Research Assistant at

    CENTEC. He is currently Project Manager for ShipsManeuvering at the Maritime Research Institute

    Netherlands (MARIN), Wageningen, The Netherlands. His research interestsare in ship maneuvering and control.

    Fernando P. Santos received the Degree and theM.Sc. degree in mechanical engineering from the In-stituto Superior Tcnico (IST), University of Lisbon,Lisbon, Portugal and from the Faculty of Scienceand Technology, Universidade Nova de Lisboa,Lisbon, Portugal, in 2002 and 2012, respectively.He also completed an Advanced Training Diplomain Risk Assessment, Safety and Reliability at IST in2007.He is a Research Assistant in the Centre forMarine

    Technology and Engineering (CENTEC), IST and isnow conducting doctoral studies on modeling and optimization of offshore windsystems reliability and maintenance.

    Miguel A. Hinostroza graduated in mechatronicsengineering from the Universidad Nacional deIngenieria (UNI), Lima, Peru, in 2012. Currently,he is working toward the M.S. degree in navalarchitecture and marine engineering at the Centre forMarine Technology and Engineering (CENTEC),Instituto Superior Tcnico, Technical University ofLisbon, Lisbon, Portugal.At CENTEC, he is working on ship dynamics and

    controls.

    Carlos Guedes Soares received the M.S. and OceanEngineer degrees from the Massachusetts Instituteof Technology, Cambridge, MA, USA, in 1976,the Ph.D. degree from the Norwegian Institute ofTechnology, Trondheim, Norway, in 1984, andthe Doctor of Science degree from the TechnicalUniversity of Lisbon, Lisbon, Portugal, in 1991.He is a Professor of Naval Architecture andMarine

    Engineering and President of the Centre for MarineTechnology and Engineering (CENTEC), a researchcenter of the Technical University of Lisbon, Lisbon,

    Portugal, which is recognized and funded by the Portuguese Foundation forScience and Technology.