A Novel Approach to Measure under Water Vehicle Disturbance

Force for Station Keeping Control

J. Manecius Selvakumar and T. Asokan

Department of Engineering Design, Indian Institute of Technology, Chennai, 600 036, India

Keywords: Under Water Vehicles, Sensor Beam, Strain Gauge, Feedback Mechanism, Disturbance Force, Station

Keeping.

Abstract: Maintenance of target position and/or orientation is essential for underwater vehicles (UWV) to successfully

complete a mission. However, in the case of work class vehicles, station keeping becomes an important

issue due to the presence of disturbance forces and requires effective feedback mechanism to maintain the

pose. Conventionally, the changes in position due to disturbance force is monitored and fed back to the

station keeping controller to make necessary corrections. This introduces unnecessary delay in response and

continuous variations in vehicle position. In this paper, an attempt has been made to develop a disturbance

force measurement setup using strain gauges which will directly measure the disturbance forces which can

be used for predicting the vehicle pose disturbance and make necessary corrections even before the vehicle

starts responding to the disturbance forces. The methodology adopted for force measurement is presented

and experimental analysis has shown promising results. This approach can be used as an alternative

feedback mechanism for station keeping control of underwater Vehicles.

1 INTRODUCTION

Station keeping is the process by which an

underwater vehicle (UWV) is held in its desired

position resisting the external forces acting on it.

Ocean medium is subjected to deep sea currents

arising due to the changes in temperature and other

factors and hence it becomes difficult to control the

undesired movements of vehicles (Eric Conrado and

Maruyama, 2007).

For station keeping control of underwater

vehicles, the feedback information about the

disturbance acting on them is very essential (Woods,

et al., 1998). However, the delay in vehicle response

and feedback cause undesired movement of vehicle

before it is controlled (Antonelli, et. al., 2001). One

way to reduce this pose disturbance is to predict the

UWV motion from the disturbance forces acting and

then generate necessary control forces to nullify the

effects of the disturbance. This will facilitate an

UWV with faster position keeping capability. In this

paper, we propose a feed-forward station keeping

control using direct measurement of disturbance

forces. Since strain gauges are widely used in

underwater applications for force measurements

(McLain and Rocky, 1992), specially designed

measurement beams with strain-gauge sensors are

proposed here for the force measurement. The

design, analysis, and experimental details of the

force sensing system are presented in this paper. The

paper is organised in the following way. The station

keeping control strategy and its simulation results

are briefly described in section 2. Section 3 and 4

describes the disturbance velocity measurement

setup. Experimentations details are covered in

section 5. The results of the sensor beam testing are

discussed in section 6.

2 CONTROL STRATEGY

The proposed control strategy is to equip the UWV

with sensors to measure the disturbance forces

acting on it and predict the resultant velocity to use

this as a feed-forward data for control. Assuming

that the disturbance forces due to underwater current

are negligible in the vertical direction, a planar

station keeping strategy is proposed. Force sensors

are attached to the UWV on all four sides and the

disturbance force is measured and sent to the

controller. Using the dynamic model of the vehicle,

the resulting velocity of the vehicle due to the

460

Manecius Selvakumar J. and Asokan T..

A Novel Approach to Measure under Water Vehicle Disturbance Force for Station Keeping Control.

DOI: 10.5220/0004036704600463

In Proceedings of the 9th International Conference on Informatics in Control, Automation and Robotics (ICINCO-2012), pages 460-463

ISBN: 978-989-8565-22-8

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

disturbance is predicted and fed to the controller as a

feed forward data. Control action is initiated by the

controller to overcome the effects of vehicle velocity

resulting from the disturbance force. Figure 1 shows

the flow chart for the above strategy. If needed, in

addition to the predicted response of the vehicle,

actual pose of the vehicle can also be used for

generating the control forces.

Figure 1: Control Scheme using Disturbance Force data.

Figure 2 shows the proposed station keeping

control methodology using the measured force. To

execute the correction in account of vehicle’s

dynamics, the conventional motion reference unit

(MRU) data is used.

Figure 2: Proposed Control Method.

The motion control algorithm based on Inverse

Dynamic Task theory (Krut’ko., 1989) is

represented in figure 3. This algorithm drives the

control function using the vehicle’s motion

parameters and is used in the station keeping control

task.

The performance of the developed control

method has been analyzed. The following is the

sample result of the simulation. Figure 4 shows the

desired position, actual position without feed

forward and with feed forward controller in x-axis.

The simulation was done for the disturbing current

Figure 3: Vehicle motion controller.

velocity which changes its magnitude and direction

at 10 sec and 40 sec. It is inferred that with the feed

forward method, the deviation from desired target

and the settling time are reduced to 50% than the

original case of performance without feed forward

loop. The simulation result reveals that the existence

of feed forward loop improves the position control

action.

Figure 4: Controller Response.

3 DISTURBANCE VELOCITY

MEASUREMENT

A prototype vehicle designed for laboratory level

experimental studies is used here for verification of

the strategy. Force sensors are arranged on the

vehicle as shown in figure 5. As a first step, the

measurement of disturbances forces arising out of

the currents is measured using this setup. The solid

model of the Experimental Underwater Vehicle

(EUWV) with its major subcomponents is shown in

figure 5.

Since measurement of disturbance force is a

major consideration in this method, the optimal

orientation and position of the sensor beam is

essential and will be decided based on the results of

the performance test of the EUWV with sensor

beam.

EUWV

MRU

A Novel Approach to Measure under Water Vehicle Disturbance Force for Station Keeping Control

461

Figure 5: Solid Model of EUWV with sensor setup.

Figure 6: Sensor Beams and Forces.

Locations of sensor beams with their significant

measurement axis and force components are

indicated in figure 6. Since, the orientation of sensor

beams F and R are perpendicular to X axis, the F

y_F

and F

y_R

are insignificant. Similarly, F

x_LF

, F

x_LR

x_RF

and F

x_RR

are also insignificant. The resultant

force in x and y axis is calculated by averaging the

force measured by the number of significant sensors

in the respective axis.

4 MEASUREMENT SETUP

The sensor beam of dimension: length (L) 300 mm,

breadth (b) 20 mm, and thickness (t) 3 mm has been

designed and used in testing.

The theoretical strain resulting from the forces

acting on the UWV is calculated as follows:

Drag force on the beam due to the current

velocity is given as (Blevins and Robert, 2003)

CAVF





(N)

(1)

where, density of water, ρ = 1000 kg/m

V – Speed of the object relative to the fluid (m/s)

A – Area of the object = L x b (m

)

– Coefficient of drag = 1.28 (Source: “Shape

Effect on Drag” at Glenn Research Centre,

NASA)

The pressure on the beam P = F

/A (N/m

) and

load per unit length w = P *b (N/m) is calculated and

the strain is calculated using Hooks law as:

ε.E = σ

(2)

where, ε is the strain acting on the beam, σ

is the

stress and E is young’s modulus of acrylic material

(Johnson and Devenport, 2007).

Confirmation of appropriate thickness, width and

the strength of the acrylic sensor beam are done by

analysis and analytical calculation. Two metal foil

strain gauges of Gauge factor (GF) 2.09 (placed on

the opposing sides and connected to half bridge

circuitry) are used for measurement of the actual

forces acting on the beam. (Hoffmann, 1889)

(Kleckers, 2004) (Schäfer, 2004).

5 TESTING OF SENSOR BEAM

Deflection members made up of Aluminium,

Stainless Steel and Acrylic were fabricated and

calibrated by the laboratory standard method

(William A. Lokos and Rick Stauf, 2004). Based on

the results of this, acrylic was chosen as the material

for sensor beam.

Figure 7: Towing Tank Facility.

Testing of the developed sensor beam was

carried out in towing tank of Ocean Engineering

Department at the Indian Institute of Technology

Madras. The carriage used for the tests has

dimension of 85 m x 3.2 m x 2.5 m and has variable

carriage speed from 0 m/s to maximum 5 m/s.

(figure 7).

In the above said towing tank testing, two

identical sensor beams as shown in Figure 8, were

fixed on the carriage and interfaced with the

instrumentation amplifier.

Data Logger PC and

Trolley Control

Panel

Electronic Housing

Thruster

Additional Buoyancy

Main Frame (PVC)

Sensor Beam

ICINCO 2012 - 9th International Conference on Informatics in Control, Automation and Robotics

462

Figure 8: Test Setup – Three Sensor Beams.

6 RESULTS AND DISCUSSION

The experiments were carried out by varying the

speed of towing trolley from 0 knot to 1 knot with

the step increase of 0.5 knot (0.25 m/s).

The result for one of the tests is shown in Figure

9, which shows the comparison of theoretical strain

and the measured strain on the 50 mm width beam

(the graphs in this section are plotted based on the

data points, however, one can easily identify the

speed increase from 0.5 knot to 1 knot and then to

1.5 knots from the graphs).

Figure 9: Results_50 mm Sensor Beam 1.

As the measured strain has constant offset from

actual, we propose to use correction factors as listed

in table 1 to use while using the output of sensor

beam for control of the EUWV.

Table 1: Correction Factors of Sensor Beams.

Width Speed

(m/s)

Correction Factor

20 mm

30 mm

40 mm

50 mm

0.25

2.5

2.05

0.5

1.25

1.4

0.75

0.625

1.025

1.125

0.825

Using the conventional relationship of force and

strain, the disturbing current velocity is derived

using the strain measured from the sensor beams.

7 CONCLUSIONS

Development of a disturbance velocity measurement

setup using strain gauges is presented in this paper.

Feed forward control strategy and simulation study

showing performance of the proposed control

scheme are presented. Study on the performance of

developed sensor setup of varying thickness and

width has been carried out. Acrylic sensor beam

having 2 mm thickness is selected to use for

implementation and experimental validation of

proposed control scheme. Derived correction factors

for the sensor beams will be used during further

experiments. This method of disturbance force

measurements can be effectively used for station

keeping control of AUVs.

REFERENCES

André Schäfer, Dr.-Ing, 2004, Force, Strain and Pressure

transducers based on foil type strain gauges, HBM

GmbH, Darmstadt, Germany.

Antonelli G., Chiaverini S., Sarkar N. & West M., 2001,

Adaptive control of an autonomous underwater

vehicle: experimental results on ODIN, In

Transactions on Control Systems Technology, IEEE,

Sep 2001.

Blevins, Robert D., 2003, Applied Fluid Dynamics

Handbook. Krieger Publishing Co.

Eric Conrado De Souza, Newton Maruyama, 2007,

Intelligent UUVs: Some Issues on ROV Dynamic

Positioning, In IEEE Transactions on Aerospace and

Electronic Systems.

Eric R. Johnson and William J. Devenport, 2007, Static

Response of a Beam, in experimental manual.

Hoffmann K., 1889, An introduction to measurements

using strain gauges, Hottinger Baldwin Messtechnik

Publisher.

Krut’ko. P. D., 1989, Inverse problems of control system

dynamics: nonlinear models, Nauka Phys. & Math

Publisher, Moscow.

McLain T. W., Rocky S. M., 1992, Experimental

Measurement of ROV Tether Tension, In Proceedings

of ROV, San Diego, CA.

Thomas Kleckers, 2004, Important characteristics of force

transducers, HBM GmbH, Darmstadt, Germany

William A. Lokos, Rick Stauf, 2004, Strain-Gauge Loads

Calibration Parametric Study, In Technical Report of

NASA Centre for Aero Space Information (CASI).

Woods A. J., Penrose J. D., Duncan A. J., Koch R., Clark

D., 1998, Improving the Operability of Remotely

Operated Vehicles, In Appea Journal, pp 849 – 854.

Strain_50 mm Sensor Beam 1

200

400

600

800

1000

1200

1400

0 500 1000 1500 2000 2500

Data Points

Strain (MicroStrain)

50_SB1_Theory

50_SB1_Experiment

50_SB1_Exp. With CF

A Novel Approach to Measure under Water Vehicle Disturbance Force for Station Keeping Control

463