DESIGN OF SERVO SYSTEM IN STATE SPACE USING

DISTANCE LABORATORY SYSTEM (DLS)

Iván Santana Ching, Luis Hernández Santana

Departamento de Automática y Sistemas Computacionales, Universidad Central “Marta Abreu” de Las Villas

Carretera a Camajuani Km. 5 ½, Santa Clara, Cuba

Manuel Ferre, Rafael Aracil

Departamento Automática, Ingeniería Electrónica e Informática Industrial

Universidad Politécnica de Madrid, Madrid, Spain

George Eisendrath

Free Universty of Brussels, Brussels, Belgium

Keywords: Distance Learning, Control Education, Remote Laboratories, Learning Management Systems.

Abstract: During last decade, the Internet has been increasingly used for education and research purposes. As

traditional face to face classroom became virtual classroom through Internet, traditional hands-on

laboratories converted as remote and virtual laboratories that are at a technological crossroad. At previous

papers was presented a Distance System Laboratory (DSL), developed for the purpose of teaching automatic

control systems. The system consists of three parts, the user interface, the practices management and the

practice processing. The developed DSL system allows users (i) learning and adjusting predefined

controllers, (ii) designing new controllers, and (iii) testing and analyzing the performance of the

predefined/designed controllers over a set of physical devices. Some technological update was made to

DLS, than increase the performance and security of the system. In this paper is described an application of

DLS through one case study practices of design of servo system in state space in the Control System

discipline.

1 INTRODUCTION

The use of simulation tools for education and

training has been increasingly becoming popular,

mainly due to the high cost of maintaining and

operating laboratories equipments. However, despite

their advantages being low cost and relatively easy

to use, the simulators do not simulate efficiently,

noise, frequency responses, D/A conversion and

other physical phenomena that characterize real

systems. (Hites, 2002).

The emergence of the Web in the mid-1990 has

added new opportunities for sharing expensive

resources of software and hardware. This allowed

the development of virtual (Anido and Fernández,

1999) and remote control (Ko et al., 2005)

applications, offering the capabilities and

flexibilities of simulation tools, without losing the

important characteristics of physical systems (Puerto

et al., 2005), and hence distance laboratory

applications. Remote experimentation facilities

offered as part of a web-based learning approach,

affords a number of critical benefits and for

engineering distance education courses it is the only

realistic method of performing many experiments.

This approach allows remotely located students to

complete laboratory assignments unconstrained by

time or geographical considerations facilitating the

development of skills in the use of real systems and

instrumentation (Callaghan et al., 2007).

In (Gravier et al., 2008), the authors provide a

literature review of modern remote laboratories and

identify possible evolutions for the next generation

of remote laboratories. Remote laboratories are

296

Santana Ching I., Hernández Santana L., Ferre M., Aracil R. and Eisendrath G. (2009).

DESIGN OF SERVO SYSTEM IN STATE SPACE USING DISTANCE LABORATORY SYSTEM (DLS).

In Proceedings of the First International Conference on Computer Supported Education, pages 296-300

DOI: 10.5220/0001971002960300

 SciTePress

under a strong current of evolution and are not

restricted to a single educational topic, where

Automatics and Robotics are ones of the most used.

In the majority of cited remote control

experiments, remote users can run an experiment

and adjust the process or controller parameters from

a set of predefined controllers. This limits the

practice to some type of controllers (PI, PID, space

of state, etc.). The Automatic Control Telelab (ACT)

enables students to choose a control law, change the

control parameters online, and even design their own

controller (Casini et al., 2007). Our developed

Distance Laboratory System (DLS) allows learning

and adjusting predefined controllers, designing new

controllers, and testing and analyzing the

performance of the predefined/designed controllers

over a set of physical devices. This has been made

possible by using Matlab-Simulink, very well known

software in the Automatic Control community

(Sartorius et al., 2005). At this stage of development,

our DSL allows several practices for the control of

direct current motor and control of a robot

manipulator. (Hernández et al., 2006).

2 CHARACTERISTICS OF THE

SYSTEM

The DLS is a distance laboratory system allowing

the users to learn how to adjust predefined

controllers and to design their own controllers in

order to test them in real devices through Internet.

2.1 Characteristics of Distance

Laboratories

DLS exhibits some features in common with most

distance laboratories in operation at present easiness,

availability and accessibility. Some additional

characteristics like easy and fast user interface,

management of multiple requests in parallel form,

controller development using MATLAB and Simulink

in a remote way, reference change are given in

(Sartorius et al., 2005). Various technological

updates were made to DLS improving the

performance and security.

The interface is based on HTML pages that use

JavaScript and now uses PHP functions. A video

feedback of the carrying out practices was

incorporate to the Web interface. The integration

with a pedagogical platform Moodle, was made too.

One of the most important features of DLS is the

management of multiple requests in parallel form.

The SLD allows to take care manifold request in a

parallel form managing centralized similar devices

that are separated geographically but connected by

(WAN).

2.2 DLS Architecture

In (Sartorius et al., 2005) is presented the DLS

architecture. That architecture was update, the CGI

located in the Web server was change for a Web

Services developed with PHP technologies. On the

other hand, in the Practice Management Client

(PMC) was introduced another Web Services that

permit communicate those parts via Web better than

TCP/IP. The actual architecture is show in Figure 1.

Figure 1: Software and hardware level architecture of

DLS.

Now, the communications port used is an

standard HTTP 80 port, rather than non-standard

TCP/IP random ports, that can be blocked by

firewalls. The system follows similar double client-

server architecture as describe in (Yan et al., 2006).

The DLS is divided into three parts: user

interface, practices management and practices

processing. The users interact with the system

through Internet. When accessing to the system Web

site, first the users must register by giving their

username and password and then choose the practice

they want to carry out. There the user can fill all the

data in the form associated to the practice in a

correct way and finally choose whether to carry out

the practice in a simulated or a real way.

The Web Service Practice Management Server

(PMS) receives the data and verifies which

workstation can carry out the practice order and

sends the order to the Web Services Practice

Management Client (PMC) located in the available

workstation. When the order arrives at the PMC, the

data are processed and the practice is carried out

using Matlab-Simulink together with Real Time

Workshop (RTW) Toolbox. Once the practice has

DESIGN OF SERVO SYSTEM IN STATE SPACE USING DISTANCE LABORATORY SYSTEM (DLS)

297

been processed, the response is a Web page showing

the processing results. Throughout the practices

carry out a video feedback with the real process is

showing to the users.

3 DISTANCE LABORATORY

SYSTEM USE

The system permits two types of practices: practices

with a predefined controller and practices with a

controller created by the user.

3.1 Practices with a Predefined

Controller

In this type of practices the users only need a Web

navigator to access the DLS Web site. This type of

page is shown in Figure 2, where it is shown a

practice for testing decoupled PID controller

performance in a robot manipulator with two

degrees of freedom. In addition, the users can

choose two execution ways: (a) a simulated way,

where it is simulated the execution and is obtained

an idealized response of the practice or (b) a real

way.

Figure 2: Web page and form for carrying out the practice

with decoupled PID controller.

3.2 Practices with a Controller Created

by the User

When the users access some practices in which they

can define the controller that they will use, it is

shown a page as the one given in Figure 3.

Figure 3: Web page and form for carrying out practice

with a controller created by the user.

From this page the user can download a Simulink

file containing the practice diagram block. For

carrying out these types of practices the users need

to have the MATLAB-Simulink software installed in

order to modify the downloaded Simulink model, as

it was explained in (Sartorius et al., 2005).

4 TEACHING EXPERIENCE

USING DLS

The DLS has been used in activities of teaching as

well as undergraduate and postgraduate course,

inside and outside our country, standing out the

accesses from Spain and Mexico, with very good

response. In (Hernández et al., 2006) was

summarized some topics, in Automatic Control

Systems education, where the developed DLS could

be effectively used.

At the next section is presented an application of

DLS through one case study practices of design of

servo system is state space in the Control System

discipline.

4.1 Design of Servo System is State

Space

The design of control system by pole-placement is

an interesting topic of Control Engineering subjects

(Ogata, 2006).

We show the practices presented for the student

in the UCLV for the design of servo system in State

Space, using the distance real laboratory (DSL). The

platform, described in the epigraph 2.2, is connected

to a real DC motor.

CSEDU 2009 - International Conference on Computer Supported Education

298

The control system implemented in the DSL has

the block diagram of figure 4.

Figure 4: Block diagram of the control system.

The motor dynamic characteristic, well know by

the student by previous work, are showed in the

figure 5. The real system has an encoder for position

measuring, and the velocity is obtained in the

computer by the derivation of encoder signal.

Figure 5: Block diagram of the motor dynamic

characteristic.

In this case the open loop control system can be

represented by:

BuAxx



(1)

Cxy =

Where x equal to state vector [

θθ



]

, motor

velocity and position.

⎥

⎦

⎤

⎢

⎣

⎡

−

67.410

00.10

⎥

⎦

⎤

⎢

⎣

⎡

65000

[]

01=C

In this case the task of the students is to design

the regulator according to performance index given

by the professors.

4.2 Student’s Task

A typical presentation of student task can have the

form:

Designs of servo system in State Space for the

system of figure 5, with closed loop response to step

input with small overshoot, 0.15 s of setting time and

zero steady-state error.

The students have several possibilities, to make

the design, according (Ogata, 2006) the servo system

pole-placement designs with state observer solution

is the best.

4.3 Student’s Task Solution

The first step is to find the analytical solution.

Following the Ogata method, (Ogata, 2006) we have

to find:

• The wanted close loop poles:

⎥

⎦

⎤

⎢

⎣

⎡

−

13.0767i 27.0000-

• The regulator gain:

[

]

0002.00138.0

• The observer gain

⎥

⎦

⎤

⎢

⎣

⎡

1.1886

0.0219

*10000Ke

With these values we implement a servo system

pole-placement designs with state observer with

extra integrator (Ogata, 2006) and test the system

solution, first in simulation in the web site with the

result of figure 6.

Figure 6: Simulation response.

DESIGN OF SERVO SYSTEM IN STATE SPACE USING DISTANCE LABORATORY SYSTEM (DLS)

299

In this case the results of simulation are

according to the analytical analysis and the

following step is test the regulator in the real motor.

The results are showed in the figure 7.

Figure 7: Real response.

5 CONCLUSIONS

In this paper have been presented an updated

Distance System Laboratory (DSL) developed for

the purpose of teaching automatic control systems.

The potential of the developed DSL system like

learning and adjusting predefined controllers and

designing new controllers have been showed. One

case of study of practices with a real motor in the

Control System discipline was presented. In the

paper is demonstrated the capacity of the system to

develop practices work in the topic of servo system

design is State Space, specifically, an example of

servo system pole-placement design with state

observer have been presented with good results.

This example demonstrated the real possibilities of

the DSL for develop real practices in this and others

topics of Automatic Control.

REFERENCES

Anido, L., Llamas, M., Fernández, M. J., 1999. SimulNet:

Building Virtual Teaching Laboratories for the Web.

In Proceedings of 8th International World Wide Web

Conference, Toronto, Canada.

Callaghan, M.J., Harkin, J., Mcginnity, T.M., Maguire,

L.P., 2007. Paradigms in Remote Experimentation. In

International Journal of Online Engineering (iJOE),

Vol. 3, No. 4.

Casini, M., Prattichizzo, D., Vicino,A., 2007. Operating

Remote Laboratories through a bootable device. In

IEEE Transactions on Industrial Electronics, vol. 54,

no. 6, December.

Gravier, C., Fayolle, J., Bayard, B., Ates, M., Lardon, J.,

2008. State of the art about remote laboratories

paradigms – foundations of ongoing mutations. In

International Journal of Online Engineering (iJOE),

Vol. 4, Issue 1, February.

Hernández, L., Sahli, H., Santana, I.., Sartorius, A., Rubio,

E., Abreu, J., 2006. Plataforma para la realización de

actividades prácticas a distancia en tiempo real.

Potencialidades de utilización en la educación

superior. In UNIVERSIDAD 2006, February.

Hites, M., 2002. Creating and Running Undergraduate

Experiments Controlled Through the Internet. In

American Society for Engineering Education,

Chicago, Illinois.

Ko, C. C., Chen, B. M., Chen, J. P., Zhang, J., Tan, K. C.,

2005. A Web-based laboratory on control of a two-

degrees-of-freedom helicopter. In Int. J. Eng. Educ.,

vol. 21, no. 6, pp. 1017–1030.

Ogata, K., 2006. Ingeniería de Control Moderna, Madrid,

España.

Puerto, R., Reinoso, O., Ñeco, R., García, N., Jiménez, L.

M., 2005. RECOLAB: Laboratorio remoto de control

utilizando Matlab y Simulink. In Revista

Iberoamericana de Automática e Informática

Industrial, vol. 2, num.2.

Sartorius, A., Hernández, L., Aracil, R., Rubio, Á.,

Santana, I.., 2005. Platform for distance development

of complex automatic control strategies using Matlab.

In The International Journal of Engineering

Education, special issue on Matlab and Simulink in

Engineering Education, Vol. 21, No. 5, pp. 790 – 797.

Yan, Y., Liang, Y., Du, X., Saliah-Hassane, H., Ghorbani,

A., 2006. Putting Labs Online with Web Services. In

IEEE Computer Society, Vol.: 8, Issue: 2, pp. 27- 34,

March-April.

CSEDU 2009 - International Conference on Computer Supported Education

300