Development of the AUV ‘ISiMI’ and a free running test in an Ocenbnan

ARTICLE IN PRESS

Ocean Engineering 36 (2009) 2–14

Contents lists available at ScienceDirect

Ocean Engineering

0029-80

doi:10.1

� Corr

E-m

journal homepage: www.elsevier.com/locate/oceaneng

Development of the AUV ‘ISiMI’ and a free running test in an OceanEngineering Basin

Bong-Huan Jun a,�, Jin-Yeong Park b, Fill-Youb Lee a, Pan-Mook Lee a, Chong-Moo Lee a, Kihun Kim a,Young-Kon Lim a, Jun-Ho Oh b

a Ocean Engineering Research Department, MOERI, KORDI, 171 Jang-dong, Yuseong-gu, Daejeon 305-343, Republic of Koreab Humanoid Robot Research Center, KAIST, 373-1 Guseong-dong, Yusong-gu, Daejeon 305-701, Republic of Korea

a r t i c l e i n f o

Article history:

Received 7 March 2008

Accepted 12 July 2008Available online 24 July 2008

Keywords:

Autonomous underwater vehicle (AUV)

Diving and steering control

Free running test

Ocean Engineering Basin (OEB)

Turning test

Zigzag test

18/$ - see front matter & 2008 Elsevier Ltd. A

016/j.oceaneng.2008.07.009

esponding author. Tel.: +82 42 868 7531; fax:

ail address: [email protected] (B.-H. Jun).

a b s t r a c t

The Korea Ocean Research and Development Institute (KORDI) has developed a small AUV named ISiMI.

The mission of ISiMI is to work as a test-bed AUV for the development and validation of various

algorithms and instruments required to enhance the AUV’s functions. The design concept of ISiMI is that

of a vehicle small enough to cruise the Ocean Engineering Basin (OEB) of KORDI and to be handled by

one or two people. The downsized design and the cruising ability in its tank enable fast experimental

feedback on AUV technologies and a shorter development period for new technologies. This paper

presents a review of our research work on the development of ISiMI, with a performance evaluation by

simulation and an experimental test. After the design and implementation of ISiMI, including its

positioning system in the OEB, are presented, a series of test results in the OEB and discussions of the

results are presented, with comparisons of the simulation and experimental outputs.

& 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Autonomous underwater vehicles (AUVs) have become a maintool for surveying below the sea in scientific, military, andcommercial applications because of the significant improvementin their performance. Despite the considerable improvement inAUV performance, however, AUV technologies are still attractiveto scientists and engineers as a challenging field. For example,multiple AUVs and underwater docking are recent challengingissues in the field of AUV technologies (Edwards et al., 2004;Fiorelli et al., 2004; Stokey et al., 2001; Singh et al., 2001). Tosuccessfully implement these new technologies in the field, anumber of sub-functions have to be tested and verified inadvance. They could be control, navigation and communicationfunctions as well as basic functions for AUVs, including anemergency architecture for survival. Since it would be veryexpensive and time consuming to conduct all these tests at sea,researchers and engineers engaged in the operation and develop-ment of underwater vehicles need easier test schemes and fasterfeedback of results in an environment similar to that of the sea.

The Maritime and Ocean Engineering Research Institute(MOERI), a branch of the Korea Ocean Research and DevelopmentInstitute (KORDI), has developed a model for an AUV named ISiMI.

ll rights reserved.

+82 42 868 7503.

ISiMI comes from the name of a small but strong fish from an oldtraditional story in Korea. In English, ISiMI is the acronym forIntegrated Submergible for Intelligent Mission Implementation.ISiMI is an AUV platform that satisfies the various needs of theexperimental tests required for the development of control andnavigational architectures or software algorithms for underwaterflying vehicles. To carry out a number of tests while avoidingmany difficulties in field trials, the first version of ISiMI wasdesigned to cruise in the Ocean Engineering Basin (OEB), a facilityin KORDI that simulates the ocean’s environment.

The main design concept of ISiMI is that of a small vehiclesmall enough to cruise in the basin and to be handled easily byone or two people. ISiMI has a Myring hull profile with a 1.2 mlength, 0.17 m diameter and 20 kg weight in air, including a 5 kgpayload. It is equipped with an attitude heading reference system(AHRS) and a depth sensor to measure its attitude, dynamicmotion and depth. A camera is mounted on its nose cone and anonboard image processing system is installed as an opticalguidance system for underwater docking. A radio frequency(R/F) modem is adopted to communicate with a surface computeron the computerized planar motion carriage (CPMC) of OEB. Theimage tracking system of CPMC is used to localize ISiMI in thebasin. The position information of ISiMI, as tracked by CPMC, istransmitted to ISiMI in real time via a R/F link.

A series of free running tests was conducted to investigate thedynamic characteristics of ISiMI and to validate its variousfunctions as a test-bed AUV. These tests included a turning test,

ARTICLE IN PRESS

B.-H. Jun et al. / Ocean Engineering 36 (2009) 2–14 3

a zigzag test, an automatic heading test, a depth control test and aguidance control test. The experimental responses were analyzedand compared with the simulation responses to verify thediscrepancies between the two sets of results.

This paper presents a review of our research work on thedevelopment of the AUV ISiMI and its performance evaluation, bysimulations and experimental tests. First, the design and im-plementation of ISiMI, including its positioning system in the OEB,will be presented. Second, a series of test results in the basin and adiscussion of the results will be presented, with comparisons ofthe simulated and experimental outputs. Finally, the second-generation ISiMI, ISiMI100, a sea-trial version of ISiMI, will beintroduced.

2. System design of the AUV

2.1. Basic design concept

The basic design concept of the AUV ISiMI is hinged on theimplementation of a small vehicle that can be easily launched,recovered and operated without special handling equipment. Thisconcept of a small AUV is intended to provide researchers withfast experimental feedback on new algorithms and instrumentsnecessary to develop AUV technologies. Several AUVs werereviewed to lead the concept of ISiMI. Sea Squirt AUV, which is1 m long and weighs 35 kg, was built in 1988 under the MIT SeaGrant to gather oceanographic data from the Charles River andserve as a test-bed for software and instrumentation, withcomponent costs of $40,000. Then the MIT Sea Grant builtOdyssey I and II for oceanographic research, with componentcosts of $50,000 and $75,000, respectively (Bellingham et al.,1993). Hydroid’s REMUS (Allen et al., 1997; Stokey et al., 2001) andGAVIA AUV, developed by WHOI and Hafmynd, respectively, aresmall commercially available AUVs with diameters less than25 cm.

MOERI-KORDI has a big square basin, OEB, for simulation ofreal sea conditions. OEB is 68.8 m long, 37.2 m wide and 4.5 mdeep. Wave, current and wind generators are installed in thebasin. The purpose of ISiMI is to serve as a test-bed AUV for thedevelopment of AUV technologies with fast experimental feed-

Fig. 2. Appearance of ISiMI with

Fig. 1. Appearance of ISiMI.

back in an OEB environment. The size of ISiMI is constrained bythe OEB environment, so ISiMI is able to run free in OEB. Itsdownsizing is at the highest level of the design process. Becausethe dimensions of the AUV conflict with the space and payload forinstruments, however, the hull size and weight of ISiMI weredetermined in the spiral design process, which is a feedbackdesign from the basic design to the detailed design.

In general, the hull shape of AUVs has mainly two types. Thefirst is the cruising type, which looks like a torpedo, and whichsurveys from hundreds of meters to hundreds of kilometers. Itsstreamlined shape is adopted for AUVs to minimize the dragforces acting on the hull while the AUV is cruising. They use axialthrusters and control planes to control their motion. The othertype of hull shape is the hovering type, which inspects a specificarea from several meters to hundreds of meters. Several thrustersare installed to keep their precise positions and attitudes inhovering motion. Since ISiMI is a test-bed for underwater surveyand docking technologies, the torpedo-type hull shape wasadopted. Moreover, a pressure hull structure was chosen ratherthan an open frame structure for spatial efficiency.

2.2. Design goals

Considering ISiMI’s basic design concept, we formulated itsdesign goals as follows:

(1)

acou

Downsizing: Its weight had to be 20–30 kg, including payloadsfor additional sensors or instruments. This scale can behandled by one or two people.

(2)
Operating speed and duration: The vehicle must have been ableto operate for more than 3 h at a speed greater than two knots.
(3)
Hull shape: The hull had to have an experimentally knownshape that would make it convenient to estimate thehydrodynamic coefficients of a mathematical model and thatwould minimize the drag force to increase the efficiency of theAUV’s thrust. It also had to have space for additionalinstruments and to endure a depth rating of more than 20 m.
(4)
Computing and processing power: The AUV’s embeddedcomputer system had to have enough computing power toprocess navigational and control algorithms and to gatherdata.
(5)
Reliability and compatibility: The system had to employ provenequipment to minimize the faults of subsystems and toincrease the reliability of the entire system. The equipmenthad to be chosen from among the vendor-supplied instru-ments continuously and stably.
(6)
Extendability: The AUV had to have the capacity to haveadditional functions with minimal changes in the basic designof its mechanical and electrical systems.
stic telemetry module.

ARTICLE IN PRESS

B.-H. Jun et al. / Ocean Engineering 36 (2009) 2–144

3. System implementation of ISiMI AUV

3.1. Overview of ISiMI

In the consideration of the design concept, a first version ofISiMI was manufactured with fundamental sensors that wererequired to enable it to cruise in the OEB. The appearance of ISiMIis shown in Fig. 1. Adopting the modular design concept, we caninsert additional mission sensors by extending the length of theAUV’s mid-section hull. Fig. 2 shows the ISiMI hull that wasextended by inserting an acoustic telemetry module, and Table 1

Fig. 3. Hull contou

Fig. 4. Conceptual view of

Table 2Hull parameters of ISiMI

Parameters Value Unit Description Remarks

a 200 mm Nose section

b 600 mm Mid-section

c 400 mm Tail section

D 170 mm Diameter

aoffset 4 mm Nose offset

boffset 30 mm Tail offset

lf 1196 mm Forward length

l 1200 mm Total length

Table 1Dimensions of ISiMI

Symbol ISiMI ISiMI100 Unit Description

L 1196 1582 mm Overall length

D 170 200 mm Diameter

W 15 38 kg Weight

U 1.0 1.5 m/s Design velocity

umax 2.0 2.0 m/s Max velocity

shows a comparison of the principal dimensions of ISiMI and theextended ISiMI. This paper focuses on the first version of ISiMIAUV to cruise in the OEB.

The ISiMI system is divided into a mechanical system, a controlsystem and a communication system. Its mechanical systemincludes its hull structure, its thruster and its control fins. Itscontrol system includes a computer, electrical interface boards,sensors and software. Its communication system, which enables itto communicate with surface or other vehicles, includes wiredand wireless LAN and a R/F modem. The extended ISiMI has anacoustic modem for underwater wireless communication. A new-generation ISiMI100 will be equipped with a hybrid navigationsystem composed of an inertial measurement unit, a Dopplervelocity log and range information (Lee et al., 2007; Lee and Jun,2007).

3.2. Mechanical system

The hull size of the ISiMI AUV is constrained by the space forthe onboard instruments, and the hull shape is constrained bythe AUV’s hydrodynamic characteristics. The standard hulllength and diameter are 1.2 and 0.17 m, respectively, consideringthe installation of the onboard equipment, which will bedescribed in the following sections. The hull shape of the ISiMIAUV was designed based on the Myring hull profile equations(Myring, 1976), which are known as the best contours forminimizing the drag coefficient for a given ratio of body lengthand diameter. The dimensions of the ISiMI AUV are shown inTable 2. Its hull contour and constitution are shown in Figs. 3and 4, respectively. Its mid-section is a pressure housing made ofaluminum, and its nose section is a flooding hull made ofpolyurethane. Its tail-section tube, which includes a mountingjig for a BLDC motor and three linear actuators, is covered by abuoyant material.

r of ISiMI AUV.

hull structure of ISiMI.

ARTICLE IN PRESS


The NACA 0012 cross-section was adopted for the controlplane. The design parameters are listed in Table 3. Linear steppermotors were chosen for the fin actuators. The pair of rudders is

Fig. 5. Control system

Fig. 6. General arrangement o

Table 4Drag force and required capacity of thrust motor for ISiMI AUV at Re ¼ 1.6E+6,

d ¼ 0.17 mm, Cd ¼ 0.2, s ¼ 1.00E+03

Speed 1 m/s 1.5 m/s 2 m/s 2.5 m/s Remarks

Fd (N) 2.2698 5.1071 9.0792 14.1867

Pd (W) 2.2698 7.6608 18.1584 35.4656

PM (W) 5.6745 19.1514 45.3960 88.6641

Table 3Fin parameters of ISiMI

Parameters Value Unit Description Remarks

Sfin 4819.5 mm2 Planform area

Xfinpost 550 mm Moment arm wrt CB

dmax 20 degree Maximum fin angle

controlled by a linear stepper motor, and the two stern elevatorsare independently driven by two linear stepper motors. Assumingthe maximum torque required by the control fin is 10 kg cm,including the torque margin for mechanical loss, and assumingthat the distance between the fin shaft and the driving axis is5 cm, we chose a 25 N stepper motor for each elevator and a 50 Nmotor for the rudders.

The capacity of the thruster motor was estimated based on thedrag equation, described as

Fd ¼ �12CdrAf ujuj, (1)

wherein Fd is the form drag acting on the hull, Cd is the dragcoefficient, Af is the maximum cross-sectional area of the hull, u isthe advanced speed, and r is the water density. The requiredpower capacity was obtained using

PM ¼ �12CdrAf ujujZu, (2)

wherein PM is the estimated power capacity of the thrust motor,and Z is the total efficiency, including the motor efficiency,the mechanical efficiency (including the friction loss) and the

diagram of ISiMI.

f the AUV ISiMI system.

ARTICLE IN PRESS

Fig. 7. Software architecture of ISiMI’s control system.

Fig. 8. Implemented hardware system of ISiMI.

Fig. 9. CPMC and image tracking system of OEB.


ARTICLE IN PRESS

Fig. 12. Free running im

Fig. 11. Launching and recovery of ISiMI in OEB.

Fig. 10. Coordinate frame.


propeller efficiency. The calculation results are shown in Table 4when the total efficiency was reduced to 0.4. To achieve morethan the maximum design speed, an 80-W BLDC motor witha reduction gear with a 5.8:1 gear ratio (made by Maxon Motors)was selected as the thruster motor. The selected propeller forthe thrust system was KP452 S175, from the KORDI propellerseries.

3.3. Control system

The control system architecture of ISiMI is shown in Fig. 5. Thestandard sensors of ISiMI include an AHRS a pressure sensor, aCCD camera and a voltage sensor to check the battery voltage. TheAHRS supplies information on the 3-axis angular velocities, 3-axisaccelerations, 2-axis inclinations and heading to the controlsystem. The depth information is gathered from the pressuresensor. A CCD camera mounted on the nose is used to detect theunderwater dock at the final stage of the underwater docking forthe terminal guidance.

The general arrangement of the parts of ISiMI is shown inFig. 6. The core of ISiMI’s control system is a single-boardcomputer interfaced with a frame grabber, a serial extensionboard and a controller area network (CAN) module via a PC104bus. A custom-made multi-purpose PCB board, which commu-nicates with a computer system via CAN, is installed to control thelinear stepper motors and the digital and analog I/O interface. Fig.5 shows a block diagram of ISiMI’s control system. The operatingsystem of the main computer is Windows XP, with real-timeextension (RTX). The application software for the graphic userinterface and the dynamic control of ISiMI is implemented withVisual C++. The software architecture of ISiMI’s control system isshown in Fig. 7.

The operating duration of ISiMI was estimated at 4 h withlithium-polymer batteries of a 207 Wh capacity, as shown inTable 4. The total weight of ISiMI in air is 20 kg, including anadditional payload of 5 kg. The implemented hardware is shownin Fig. 8.

age of ISiMI AUV.

ARTICLE IN PRESS


3.4. Communication system

A wireless local area network (LAN) was adopted as thecommunication system between ISiMI and a surface computer,which is used for offline communication for the mission allocationand data downloading. A wired LAN is additionally used asbackup. A R/F modem is installed for online communicationbetween ISiMI and the surface computer for real-time dataexchange. Generally, R/F cannot be used underwater because ofthe severe attenuation. We experimentally confirmed, however,that it is practical to transmit packets bi-directionally up to 3.5 mdeep in the OEB with an R/F modem with a bandwidth of151.3 MHz and an output of 10 mW. With the R/F link, the user’scommands and ISiMI’s position are transmitted from the surfacePC to ISiMI, and acknowledgments of reception are returned to thesurface PC.

Fig. 14. Experimental results

0 200 400 600 800 1000 1200 1400 1600 18000

0.5

1

1.5

2

2.5

thruster rpm

velo

city

[m/s

ec]

Fig. 13. Advance speed test of ISiMI.

3.5. Localization in OEB

The localization problem is a major challenge in underwaterrobotics. In the OEB environment, there is a severe constraint tousing an acoustic positioning system due to the multi-path of theacoustic signal reflected from the wall and bottom of the OEB andfrom the free surface. Therefore, the image tracking system on theCPMC in the OEB is used for the non-contract position tracking ofISiMI. The CPMC has a three-degrees-of-freedom moving mechan-ism that enables ISiMI to follow the x–y position and heading of anobject located under the CCD camera. The position and heading ofthe object with respect to the CCD coordinates are transformed tothe reference coordinates (the basin coordinates) using theposition and heading of the camera with respect to the referencecoordinates, which are calculated from the encoders of the CPMCactuators. The tracked position of ISiMI is transmitted to the ISiMIwith a 2 Hz bandwidth via a wireless R/F link between the surfacecomputer and ISiMI in real time. The image tracking systeminstalled under CPMC is shown in Fig. 9.

4. Numerical model and controller design

4.1. Numerical model

The three-dimensional non-linear dynamic equation of sub-mersible motion has been described by Gertler and Hagen (1967),Feldman (1979) and Fossen (1994). Ignoring the effect of current,we can describe the dynamic equation of ISiMI as the form in thestudy of Feldman (1979) or Gertler and Hagen (1967). Consideringthe body coordinate system in Fig. 10 and classifying the terms inthe dynamic equation as in Jeon et al. (2003), we get the followingequation for ISiMI:

M _v ¼ FCC þ Fvh þ Frest þ Fthrust þ Ffin, (3)

wherein v ¼ {u,v,w,p,q,r}T is the linear and angular velocity vectorwith respect to the body coordinate frame, M is the inertial termincluding the added mass, FCC is the coriolis and centrifugal force

of open-loop zigzag test.

ARTICLE IN PRESS

5


term for a rigid body, Fvh is the velocity-dependent hydrodynamicforce acting on the body, Frest is the restoring force, Fthrust is thethrust force and Ffin is the lift and drag force on the fins. Eachelement term in (3) is listed in Appendix A. The hydrodynamiccoefficients in the model were estimated using the Nernstein andSmith method and the Prestero method (Nerstein and Smith,1968; Prestero, 2001). The developed coefficients were non-dimensionalized with the length of the vehicle and are listed inAppendix B. The roll damping coefficient was analogized fromthat of similar vehicles.

The velocity v in (3) can be written with respect to the earth-fixed coordinate frame with the transformation matrices J’s asfollows:

_g1 ¼ J1ðg2Þv13v1 ¼ J�11 ðg2Þ _g1, (4)

and

_g2 ¼ J2ðg2Þv23v2 ¼ J�12 ðg2Þ _g2, (5)

wherein v1 ¼ {u,v,w}T and v2 ¼ {p,q,r}T are the linear and angularvelocities with respect to the body coordinate frame, respectively;g1 ¼ {X,Y,Z}T and g2 ¼ {f,y,c}T are the position and attitudevectors with respect to the earth-fixed frame, respectively; andJ1, J2 are the linear and angular velocity transformation matricesreferred to in Appendix A. Based on the non-linear dynamics ofISiMI, we developed a simulation environment for it using MatLaband Simulink. All the simulation results presented in this paperwere derived from the simulation environment.

-18 -16 -14 -12 -10 -8 -6 -4 -2 0-10

-5

0X: -13.67Y: -2.01

Pos X (m)

Pos

Y (m

)

X: -1.36Y: -1.86

Steady turningradius :6m

0 20 40 60 80 100 120 140 160-10

-5

0

5

10

15

time [sec]

Ang

ular

Vel

ocity

(deg

/s)

yaw rateδr

0 20 40 60 80 100 120 140 1600

0.2

0.4

0.6

0.8

time [sec]

spee

d[m

/sec

]

Fig. 15. Experimental results of open-loop turning test: (a) turning radius; (b)

steady state turning rate and (c) steady state turning speed.

4.2. Controller design

The sliding mode controller has been successfully applied tothe control of underwater vehicles (Healey and Lienard, 1993;Utkin, 1977; Lee et al., 1999). The sliding mode controller was usedas the motion controller of ISiMI. The simplified equations for thesteering and diving motions can be derived from (A.1) by ignoringthe cross-flow terms and higher-order terms. The linearizedsteering system dynamics are given by

_xhðtÞ ¼ AhxhðtÞ þ bhdRðtÞ, (6)

Ah ¼Iz �

r2

l5N_r 0

0 1

24

35�1 r

2l4Nr 0

1 0

24

35,

bh ¼Iz �

r2

l5N_r 0

0 1

24

35�1 r

2l3U2NdR

0

24

35,

xh ¼ ½ r c �T.

Substituting the coefficients in Appendix B into (6) and the slidingpoles of the steering system arbitrarily at [0–2], we get thesteering control law as

dR ¼ �0:0281r þ 2:58Z tanh ðr þ 2ðc�ccomÞ=1Þ. (7)

The linearized diving system dynamics are given by the third-order system

_xvðtÞ ¼ AvxvðtÞ þ bvdSðtÞ (8)

Av ¼

Iy �r2

l5M _q 0 0

0 1 0

0 0 1

2664

3775�1 r

2l4Mq �zGW 0

1 0 0

0 �u 0

2664

3775,

bv ¼

Iy �r2

l5M _q 0 0

0 1 0

0 0 1

2664

3775�1 r

2l3U2NdS

0

0

2664

3775,

xv ¼ ½ q y z �T.

Substituting the coefficients of vertical motion in Appendix B andplacing the sliding poles of the system at [0�2.5�2.4], we get thediving control law as

dS ¼ 0:4414qþ 0:5309yþ 4Z tanh ðqþ 4y� 4ðz� zcomÞ=0:65Þ.

(9)

The line of sight (LOS) is the horizontal plane angle for theguidance function of ISiMI. It was derived from

ccom ¼ tan�1 ðYk � YðtÞÞ

ðXk � XðtÞÞ

� �, (10)

ARTICLE IN PRESS


in which ½Xk Yk � are the waypoints preprogrammed in thevehicle, and ½XðtÞ YðtÞ � is the current location of the vehicle. Thedecision as to whether or not the waypoint has been reached ismade from

r2ðtÞ ¼ ½Yk � YðtÞ�2 þ ½Xk � XðtÞ�2or20; 0olo1. (11)

Fig. 16. Experimental results of closed-loop de

Fig. 17. Experimental result

5. Free running tests in the OEB

A series of free running tests of ISiMI was carried out in theOEB. The test was intended to determine ISiMI’s maneuveringcharacteristics and to validate its basic functions as a test-bedAUV. The results might be used as references for the design of

pth control with sliding mode controller.

s of waypoint tracking.

ARTICLE IN PRESS

Fig. 18. Sea-trial version of ISiMI AUV named ISiMI100.


high-level control algorithms such as path planning or of theguidance system of ISiMI. ISiMI was easily launched without anyspecial device, as shown in Fig. 11. The image tracking system ofCPMC gathered the velocity and position of ISiMI during the freerunning test. Fig. 12 shows the free running test of ISiMI trackedwith CPMC.

5.1. Advanced speed test

An advanced speed test was carried out, and the fulfillment ofthe design speed was verified. The advanced speed of ISiMI withrespect to the propeller’s rpm was determined, as shown in Fig. 13.The design speed (1 m/s) was reached at a propeller speed ofabout 1000 rpm. By extrapolating the results shown in Fig. 13, it isestimated that the maximum speed (2 m/s) was reached at1650 rpm.

5.2. Open-loop zigzag test

The zigzag test was conducted to check the horizontalmaneuvering property of ISiMI. The results of the test in thehorizontal plane were plotted with the simulation resultsconducted with the 6-DOF non-linear model in Fig. 14. Thevelocity of ISiMI was 0.8 m/s and its rudder angle was toggledbetween �12.61 and +12.61. The experimental results showed thatthe overshoot angle was in the range of 5–7.61. The hydrodynamiccoefficients used in the simulation are listed in Appendix B. Thediscrepancy between the simulation and experiment resultsshows that the numerical model has more yaw damping or lessfin force than ISiMI. The experimental results will be used insystem identification for the correction of the numerical model infuture works.

5.3. Open-loop turning test

To calculate ISiMI’s turning radius, a turning test was carriedout. The test results are plotted in Fig. 15, from which the turningcharacteristics of ISiMI were analyzed. The steady turning radiuswas about 6 m, and the steady turning speed and rate were about0.6 m/s and �61/s when the rudder angle was 151 at the advancedspeed of 0.7 m/s.

5.4. Closed-loop depth control

As the basic functions of an AUV test-bed, the depth controland waypoint tracking control functions were tested. A slidingmode controller was designed and tested for ISiMI’s depth control.The experimental results are plotted with the simulation results ofthe linear and non-linear models in Fig. 16. The initial depth was0.4 m, and the reference depth was 1 m. Both the experimental

and simulated results showed good convergence with thereference depth. The settling time of the experiment results waslonger, however, than that of the simulation results, and theamplitude of the pitch rate response in the simulation was largerthan that in the experiment results. It is supposed that thediscrepancies in the responses are due to the errors of thenumerical and real model in pitch motion. A more exact numericalmodel will be estimated based on the experimental data using thesystem identification method in future works.

5.5. Waypoint tracking

ISiMI was guided to track a figure-eight trajectory to test theLOS algorithm described in Section 4.2. The position of ISiMIwas measured using the image tracking system in CPMC, andtransmitted to ISiMI via the R/F link. The position of the waypointswas pre-recorded in ISiMI’s memory. The threshold level r0 in (11)was 1 m. The results are plotted in Fig. 17. The yaw angle in Fig. 17was controlled by the PD controller to follow the yaw referencegenerated by the LOS. It was verified that waypoint tracking, thebasic function of the test-bed, was successfully achieved.

6. Conclusion

In this paper, the design, implementation and test results of asmall AUV named ISiMI are presented. The AUV, ISiMI, developedin KORDI is a test-bed for the validation of the algorithms andinstruments of the AUV. For fast experimental feedback on newalgorithms, ISiMI was designed to be able to cruise in the OceanEngineering Basin environment at KORDI. The zigzag test and theturning test were carried out to check ISiMI’s maneuveringproperties. The depth control and waypoint tracking tests werecarried out to validate the feedback controller of ISiMI. Theexperiment results were compared with those of the simulation.The research works were fed back to the design and implementa-tion of a 100-m-class AUV named ISiMI100. ISiMI100 is equippedwith additional sensors such as a Doppler velocity log, an acoustictelemetry modem, an obstacle avoidance sonar, a range sonar anda GPS module. A photo of ISiMI100 is shown in Fig. 18. The missiontest of ISiMI and the sea-trial of ISiMI100 remain to be performedin future works.

Acknowledgments

This work was supported in part by MLTMA of Korea for the‘‘development of a deep-sea unmanned underwater vehicle,’’ andKORDI, for the ‘‘development of ubiquitous-based key technolo-gies for the smart operation of maritime exploration fleets.’’

ARTICLE IN PRESS


Appendix A. Element terms of dynamic equations of motion

�
Inertial matrix
M ¼

m�r2

l3X _u 0 0 0 mzg �myg

0 m�r2

l3Y _v 0 �mzg �r2

l4Y _p 0 mxg �r2

l4Y _r

0 0 m�r2

l3Z _w myg �mxg �r2l5Z _q 0

0 �mzg �r2

l4K _v myg Ix �r2

l5K _p �Ixy �Izx �r2

l5K _r

mzg 0 �mxg �r2

l4M _w �Ixy Iy �r2

l5M _q �Iyz

�myg mxg �r2

l4Y _r 0 �Izx �r2

l5K _r �Iyz Iz �r2

l5N_r

2666666666666666664

3777777777777777775

. (A.1)

�
Coriolis and centrifugal force of a rigid body:
FCC ¼ CRBðvÞv ¼ ½XCC YCC ZCC KCC MCC NCC �T, (A.2)

XCC ¼ mvr �mqw�mygpq�mzgpr þmxgq2 þmxgr2,

YCC ¼ �mur þmwp�mxgpq�mzgqr þmygp2 þmygr2,

ZCC ¼ muq�mvpþmxgrp�mygqr þmzgp2 þmzgq2,

KCC ¼ myguqþmzgur �mygvp�mzgwp� ðIz � IyÞqr þ Iyxq2 � Iyxr2 þ Ixzpq� Ixypr,

MCC ¼ mzgvr þmxgvp�mxguq�mzgwq� ðIx � IzÞrp� Ixzp2 þ Ixzr2 � Iyzpqþ Ixyqr,

NCC ¼ �mxgur �mygvr þmygwqþmxgwp� ðIy � IxÞpqþ Ixyp2 � Ixyq2 þ Iyzrp� Izxqr.

�
Velocity-dependent hydrodynamic force:
FDL ¼ ½XDL YDL ZDL KDL MDL NDL �T, (A.3)

XDL ¼r2

l2ðXvvv2 þ Xwww2Þ þr2

l4ðXqjqjq2 þ Xrrr

2Þ,

YDL ¼r2

l2ðYvvþ YvjvjvjvjÞ þr2

l3Yrr þr2

l4Yrjrjrjrj,

ZDL ¼r2

l2ðZvvv2 þ Zwwþ ZwjwjwjwjÞ þr2

l3Zqqþr2

l4Zqjqjqjqj,

KDL ¼r2

l3Kvvþr2

l4Krr þr2

l5Krjrrjrj,

MDL ¼r2

l3ðMvvv2 þMwwþMwjwjwjwjÞ þr2

l4Mqqþr2

l5Mqjqjqjqj,

NDL ¼r2

l3ðNvvþ NvjvjvjvjÞ þr2

l4Nrr þr2

l5Nrjrjrjrj.

�
Restoring force:
Frest ¼

Xrest

Yrest

Zrest

Krest

Mrest

Nrest

26666666664

37777777775¼

�ðW � BÞsyðW � BÞcysfðW � BÞcycf

ðygW � ybBÞcycf� ðzgW � zbBÞcysf�ðxgW � xbBÞcycf� ðzgW � zbBÞsyðxgW � xbBÞcysfþ ðygW � ybBÞsy

26666666664

37777777775

(A.4)

�
Thrust force
Fthrust ¼ ½Xthrust Y thrust Zthrust Kthrust Mthrust Nthrust �T, (A.5)

Xthrust ¼r2

l2ðaX þ bXuT þ cXu2T þ dXu3

TÞ,

Y thrust ¼ Zthrust ¼ Mthrust ¼ Nthrust ¼ 0,

Kthrust ¼r2

l2ðaK þ bKuT þ cKu2T þ dKu3

TÞ,

and

uT ¼u0n

n0u� 1,

wherein n0, u0 are the designed rps of the propeller and the designed forward speed of the vehicle, respectively.

ARTICLE IN PRESS


�
Fin force:
Ffin ¼ ½Xfin Y fin Zfin K fin Mfin Nfin �T, (A.6)

Xfin ¼r2

l3U2XdRdR,

Y fin ¼r2

l3U2YdRdR,

Zfin ¼r2

l3U2ZdSdS,

K fin ¼r2

l3U2KdSSdSS �

r2

l3U2KdSPdSP,

Mfin ¼r2

l3U2MdSdS,

Nfin ¼r2

l3U2NdRdR.

�
Coordinates transformation matrix:
J1ðg2Þ ¼

cos c cos y � sin c cos fþ cos c sin y sin f sin c sin fþ cos c cos f sin ysin c sin y cos c cos yþ sin f sin y sin c � cos c cos fþ sin y sin c cos f� sin y cos y sin f cos y cos f

264

375; J2ðg2Þ ¼

1 sin f tan y cosf tan y0 cos f � sin f0 sin f= cos y cos f= cos y

264

375.

(A.7)

Appendix B. Hydrodynamic and thrust coefficients of the ISiMI

m ¼ 21.8 kg
D ¼ 213.64 N W ¼ 213.64 N l ¼ 1.2 m
Ix ¼ 0.1821
Iy ¼ 1.8207 Iz ¼ 1.8207
Ixy ¼ Iyz ¼ Izx ¼ 0

u0 ¼ 1
n0 ¼ 5.2
(xg, yg, zg) ¼ (0, 0, 0.03)
(xb, yb, zb) ¼ (0, 0, 0)
Hydrodynamic coefficients:

X _u ¼ �0:000887

Xvv ¼ �0.005991
Xww ¼ �0.005991
Xqq ¼ �0.002283
Xrr ¼ �0.002283
Yv ¼ �0.025040
Yv ¼ �0.062586 Yv|v| ¼ �0.135039
Y _r ¼ 0:000660
Yr ¼ 0.016100 Yr|r| ¼ �0.003567
Zvv ¼ 0

Zw ¼ �0.025040
Zw ¼ �0.062586 Zw|w| ¼ �0.135039
Z _q ¼ �0:000660
Zq ¼ �0.016100 Zq|q| ¼ 0.003567
Kv ¼ 0
Kv ¼ 0
K _p ¼ �0:00006
Kp ¼ �0.000005
Kr ¼ 0
Kr ¼ 0 Kr|r| ¼ 0
Mvv ¼ 0

M _w ¼ �0:000660
Mw ¼ 0.030853 Mw|w| ¼ 0.011760
M _q ¼ �0:001502
Mq ¼ �0.007543 Mq|q| ¼ �0.001563
Nv ¼ 0.000660
Nv ¼ �0.030853 or 0 Nv|v| ¼ �0.011760
N_r ¼ �0:001502
Nr ¼ �0.007543 Nr|r| ¼ �0.001563
XdR¼ 0
YdR
¼ �0:043008
ZdS¼ �0:043008
KdSS¼ KdSP

¼ 0
MdS¼ �0:018192 NdR
¼ �0:018192

Thrust coefficients:

aX ¼ 0
bX ¼ 0.01802 cX ¼ 0.0012 dX ¼ 0.0002405
aK ¼ 0
bK ¼ �0.000236 cK ¼ 0.000306 dK ¼ 0.000084
References

Allen, B., Stokey, R., Austin, T., Forrester, N., Goldsborough, R., Purcell, M., von Alt, C.,1997. REMUS: a small, low cost AUV; system description, field trials andperformance results. Proceedings of the Oceans Conference, 994–1000.

Bellingham, J.G., Bales, J.W., Goudey, C.A., Consi, T.R., 1993. Performancecharacteristics of the Odyssey AUV. In: Proceedings of the Eighth InternationalSymposium on Unmanned Untethered Submersible Technology (AUV ‘93),pp. 37–59.

Edwards, D.B., Bean, T.A., Odell, D.L., Anderson, M.J., 2004. A leader–followeralgorithm for multiple AUV formations. In: Proceedings of IEEE/OES, pp. 40–46.

Feldman, J., 1979. DTNSRDC revised standard submarine equations of motion.DTNSRDC/SPD-0393-09.

Fiorelli, E., Leonard, N.E., Bhatta, P., Paley, D., Bachmayer, R., Fratantoni, D.M., 2004.

Multi-AUV control and adaptive sampling in Moterey Bay. Workshop on

Multiple AUV Operations (AUV04), pp. 134–147.

Fossen, T.I., 1994. Guidance and Control of Ocean Vehicles. Wiley, New York.

Gertler, M., Hagen, G.R., 1967. Standard equations of motion for submarine

simulations. NSRDC Report No. 2510.

Healey, A.J., Lienard, D., 1993. Multivariable sliding-mode control for autonomous

diving and steering of unmanned underwater vehicles. IEEE Journal of Oceanic

Engineering 18 (3), 327–339.

Jeon, B.H., Lee, P.M., Li, J.H., Hong, S.W., Kim, Y.G., Lee, J., 2003. Multivariable

optimal control of an autonomous underwater vehicle for steering and diving

control in variable speed. In: Proceedings of Oceans Conference, San Diego,

pp. 2659–2664.

ARTICLE IN PRESS


Lee, P.M., Jun, B.H., 2007. Pseudo long base line navigation algorithm forunderwater vehicles with inertial sensors and two acoustic range measure-ments. Ocean Engineering 34 (3–4), 416–425.

Lee, P.M., Hong, S.W., Lim, Y.K., Lee, C.M., Jeon, B.H., Park, J.W., 1999. Discrete-timequasi-sliding mode control of an autonomous underwater vehicle. IEEE Journalof Oceanic Engineering 24 (3), 388–395.

Lee, P.M., Jun, B.H., Kim, K.H., Lee, J.H., Aoki, T., Hyakudome, T., 2007. Simulation ofan inertial acoustic navigation system with range aiding for an autonomousunderwater vehicle. IEEE Journal of Oceanic Engineering 32 (2), 329–345.

Myring, D.F., 1976. A theoretical study of body drag in subcritical axisymmetricflow. Aeronautical Quarterly 27, 186–194.

Nerstein, W., Smith, K.C., 1968. Hydrodynamic coefficient equations and computerprograms. The Bendix Corporation Report.

Prestero, T., 2001. Verification of a six-degree of freedom simulation model for theREMUS autonomous underwater vehicle. M.S. Dissertation, MIT and WHOI.

Singh, H., Bellingham, J.G., Hover, F., Lerner, S., Moran, B.A., Heydt, K., Yoerger, D.,2001. Docking for an autonomous ocean sampling network. IEEE Journal ofOceanic Engineering 26 (4), 498–514.

Stokey, R., Allen, B., Austin, T., Goldsborough, R., Forrester, N., Purcell, M., Alt, C.V.,2001. Enabling technologies for REMUS docking: an integral component of anautonomous ocean-sampling network. IEEE Journal of Oceanic Engineering 26(4), 487–497.

Utkin, V.I., 1977. Variable structure system with sliding modes. IEEE Transactionson Automatic Control 22 (2), 212–222 /http://auvlab.mit.edu/vehicles/vehiclespecEARLY.html#OD1S /http://www.gavia.is/downloads/brochures/GaviaBrochure0402.pdfS.

Documents

Development of the AUV ‘ISiMI’ and a free running test in an Ocenbnan