Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

Product Lifecycle Management and Embedded Information Devices References 765

43.17 I. Inasaki, H.K. Tönshoff: Roles of sensors in man-

ufacturing and application ranges. In: Sensors in

Manufacturing, ed. by H.K. Tönshoff, I. Inasaki

(Wiley, New York 2001)

43.18 CIMdata: Product life cycle management – em-

powering the future of business, Technical Report

(CIMdata, 2002)

43.19 H. Cao, P. Folan, L. Zheng Lu, J. Mascolo,

N. Frantone, J. Browne: Design of an end-of-life

decision support system using product embed-

ded information device technology, ICE Conf. Proc.

(2006)

43.20 PROMISE: PROMISE case studies, public version

available at www.promise-plm.com

Part E 43

“This page left intentionally blank.”

767

Education an

44. Education and Qualiﬁcation for Control

and Automation

Bozenna Pasik-Duncan, Matthew Verleger

Engineering education has seen an explosion of

interest in recent years, fueled simultaneously by

reports from both industry and academia. Au-

tomatic control education has recently become

a core issue for the international control com-

munity. This has occurred in tandem with the

explosion of interest in engineering education as

a whole. The applications of control are growing

rapidly. There is an increasing interest in con-

trol from researchers from outside of traditionally

control-based ﬁelds such as aeronautics, chemical,

mechanical, and electrical engineering. Recently

control and systems theory have had much to

offer to nontraditional control ﬁelds such as bi-

ology, biomedicine, ﬁnance, actuarial science,

and the social sciences as well as transportation

and telecommunications networks. Complemen-

tary, innovative developments of control and

systems theory have been motivated and inspired

by complex real-world problems. These new de-

velopments present huge challenges in control

education. Meeting these challenges will require

a multifaceted approach by the control community

that includes new approaches to teaching, new

preparations for facing new theoretical control and

systems theory problems, and a critical review of

the status quo. This chapter discusses these new

challenges as well as new approaches to education

and outreach. This chapter starts by presenting an

argument towards the future of controls as the

application of control theory expands into new

and unique disciplines. It provides two case stud-

ies of nontraditional areas where control theory

has been applied: ﬁnance and biomedicine. These

two case studies show a high potential for us-

ing powerful fundamental principles and tools of

automatic control in research with an inter-

44.1 The Importance of Automatic Control

in the 21st Century................................ 768

44.2 New Challenges for Education................ 768

44.3 Interdisciplinary Nature

of Stochastic Control ............................. 769

44.4 New Applications of Systems

and Control Theory ............................... 770

44.4.1 Financial Engineering

and Financial Mathematics .......... 770

44.4.2 Biomedical Models: Epilepsy Model 771

44.5 Pedagogical Approaches........................ 772

44.5.1 Coursework ................................ 772

44.5.2 Laboratories as Interactive

Learning Environments ................ 773

44.5.3 Plain Talk on Control

for a Wide Range of the Public...... 774

44.5.4 New Approaches to Cultivating

Students Interest in Math, Science,

Engineering, and Technology

at K-12 Level............................... 774

44.6 Integrating Scholarship, Teaching,

and Learning ....................................... 775

44.7 The Scholarship of Teaching and Learning 775

44.8 Conclusions and Emerging Challenges .... 776

References .................................................. 776

disciplinary nature. The chapter then outlines cur-

rent and future pedagogical approaches being

employed in control education, particularly intro-

ductory courses, around the world. It concludes

with a discussion about the role of scholarship,

teaching, and learning in control education both

now and in the coming years.

Part E 44

768 Part E Automation Management

44.1 The Importance of Automatic Control in the 21st Century

The ﬁeld of automatic control has a rich history

(Fig.44.1), well presented in a special issue of the Euro-

pean Journal of Control [44.1]. Fleming[44.2] provides

an assessment of its status and needs of control theory

through the 1980s. A broader and more updated report

is provided in [44.3–5].

At its core, control systems engineering involves

a variety of tasks including modeling, identiﬁca-

tion, estimation, simulation, planning, decision making,

optimization, and deterministic and stochastic adap-

tation. While the overarching purpose of any con-

trol system is to assist with the automation of an

event, the successful application of control princi-

ples involves the integration of various tools from

related disciplines such as signal processing, ﬁl-

tering, stochastic analysis, electronics, communica-

tion, software, algorithms, real-time computing, sen-

sors and actuators, as well as application-speciﬁc

knowledge.

a) b)

Fig. 44.1a,b The centrifugal governor (a) is widely considered to be the ﬁrst practical control system, dating back to the

1780s. It was used in the Boulton and Watt steam engine

(b) to regulate the amount of steam allowed into the cylinders

to maintain a constant engine speed

The applications of control automation range from

transportation, telecommunications networks, manufac-

turing, communications, aerospace, process industries

through commercial products reaching as diverse ﬁelds

of study as biology [44.6], medicine [44.7–10], and ﬁ-

nance [44.11, 12]. New issues are already appearing in

thenextgenerationoftransportation[44.13]andtelecom-

munications network [44.14] problems, as well as in the

next generation of sensor networks (particularly for ap-

plications such as weather prediction), emergency re-

sponsesystems,andmedicaldevices.Thesearenewchal-

lenges thatcan be solved through the careful application

ofcontrolandsystemstheory.Theimpactofsystemsand

control in the changing world is described in the 2007

ChineseControlConference(CCC)plenarytalkSystems

and Control Impact in a Changing World delivered by

Ted Djaferis, the 2007 President of the Control Systems

Society [44.15]. A more thorough description of control

theory forautomation canbe foundin Chaps. 9–11.

44.2 New Challenges for Education

Marketplace pressures and advances in technology are

driving a need in modern industry for well-trained con-

trol and systems scientists. With its cross-boundary

nature and ever-growing application base, helping

students in all disciplines of science, technology, en-

gineering, and mathematics (STEM) to understand the

power and complexity of control systems is becoming

an even more critical component to the future of tech-

nical education. The need to train all STEM graduates

to be comfortable with control theory generates many

Part E 44.2

Education and Qualiﬁcation for Control and Automation 44.3 Interdisciplinary Nature of Stochastic Control 769

new challenges in control education which are exten-

sively discussed in [44.16]aswellasinthereportof

the National Science Foundation (NSF) and the Control

Systems Society (CSS) panel on an assessment of the

ﬁeld [44.4] with its summary given in [44.5].

While the skills necessary for students to become

successful practitioners of their craft are changing, so

too is the background of our students. They are better

prepared to work with modern computing technologies.

The ability to interact with and manipulate a computer

is second nature to today’s connected student [44.18].

Thus the time is ripe for major renovations in con-

trol education as it applies to STEM disciplines. The

ﬁrst step in the renovation process is to develop cross-

disciplinary examples, demonstrations, and laboratory

exercises that illustrate systems and control across the

entire spectrum of STEM education [44.3–5,19].

The recent National Academy of Engineering

(NAE) report [44.17] identiﬁed the attributes and

abilities engineers will need to perform well in

a world driven by rapid technological advancements,

national security needs, aging infrastructure in devel-

oped countries (Table 44.1), environmental challenges

brought about by population growth and diminish-

ing resources, and creation of new disciplines at

the interfaces between engineering and science. The

systems and control community has been actively

involved and engaged in taking a leading role in

shaping the future automatic control engineering cur-

Table 44.1 America’s infrastructure is in dire need of re-

pair: just one of the many problems today’s engineering

students will be facing as they enter the workforce (af-

ter [44.17])

Area Grade Trend

(since 2001)

Roads D

↓

Bridges C ↔

Transit C ↓

Aviation D ↔

Schools D ↔

Drinking water D ↓

Watewater D ↓

Dams D ↓

Solid waste C

↔

Hazardous waste D

↔

Navigable waterways D

↓

Energy D

↓

America’s infrastructure D

GPA

total US$ 1.6 trillion

investement

(estimated 5-year

need)

grade point average

riculum [44.20], as well as in educating and making

nonengineering communities aware of the beneﬁts and

the power of the systems and control approaches and

tools.

44.3 Interdisciplinary Nature of Stochastic Control

Stochastic adaptive control, whereby the unknown pa-

rameters of a control system are modeled as random

variable or random processes [44.21, 22], can be used

to illustrate the interdisciplinary nature of control.

The general approach to adaptive control (Fig.44.2)

involves a splitting, or separation, of parameter iden-

tiﬁcation and adaptive control. A system’s behavior

depends on some set of parameters, and the fact that the

values of the parameters are unknown makes the sys-

tem unknown. Some crucial information concerning the

system is not available to the controller, and this infor-

mation can be learned during the system’s performance.

Using that information, the system’s performance can

then be altered to respond. This altered response may in

turn alter the previously unknown parameters. This pro-

cess is repeated in a recursive manner until the system

is shut down.

The described problem is the basic problem of

adaptive control. A stochastic control system can be

described using a stochastic differential equation or

a stochastic partial differential equation. The solution

to the adaptive control problem consists of showing the

strong consistency of the family of estimators of an un-

known parameter and the self-optimality of an adaptive

control that uses the family of estimates. The distur-

bance, or noise, in the system is modeled by a Brownian

motion or more generally by a fractional Brownian

motion (more accurate for recent problems in telecom-

munication, ﬁnance or biomedicine) [44.11]. Industrial

operation models are often described by stochastic con-

trolled systems [44.23].

Let us describe a simple adaptive control using

a simple investment model. Consider a model where an

investor has a choice in investing in two assets, a simple

Part E 44.3

770 Part E Automation Management

System

Generic

system

operations

System

output

Parameter

identification

Unknown

control

parameters

Control

parameter

values

Adaptive

control

savings account with a ﬁxed rate of growth (say 3.5%)

or shares of Microsoft stock whose growth is governed

by a Brownian motion with unknown drift and variance.

The investor controls his asset by transferring money

between the stock and the savings account. The goal

is, when the stock is expected to go down, to sell the

Fig. 44.2 The general structure of an adaptive control sys-

tem is one where the critical input control parameters are

outputs of thesystem. Asthe systemoperates overtime, the

parameters are adaptedto control how thesystem functions

in the future



stock and transfer the money to the savings account, but

when the stock is expected to go up to repurchase the

stock before it goes up. The control variable is the to-

tal amount of money transferred from the stock to the

savings account and can be positive (stock has just been

sold) or negative (stock is about to be purchased). The

savings account is governed by a traditional differen-

tial equation and the stock is governed by a stochastic

differentialequation. Thegoal is toﬁnd the optimalcon-

trol so that the expected rate of growth is maximized.

The identiﬁcation problem is to estimate the unknown

parameters, the drift and variance of the stock’s random-

ness, based on the available observations. The adaptive

control problem is to construct the (certainty equiv-

alence) adaptive control as a function of the current

state and the current estimate of what is expected to

happen.

44.4 New Applications of Systems and Control Theory

The use of systems and control theory has seen an ex-

pansion in recent years into new and emerging ﬁelds of

study. This expansion functions as a control system in

itself, with the unknown parameter being the number of

STEM students receiving controls training. The expan-

sion into new disciplines demonstrates the necessity of

controls education for all students in STEM disciplines

by showing that controls theory can be applied in a wide

varietyof unique places for producingsolutions to some

of today’s most complex problems. This demonstration

in turn acts as a driver for the curricular change nec-

essary to result in more STEM students experiencing

control education. As those students graduate, they too

will push the boundaries of what control theory can be

applied to and in turn feed the expansion cycle.

44.4.1 Financial Engineering

and Financial Mathematics

Research in control methods in ﬁnance has experi-

enced tremendous progress in recent years. The rapid

progress has necessitated communication and network-

ing between researchers in different disciplines. This

area brings together researchers from mathematical sci-

ences, ﬁnance, economics, and engineering. Together,

these people work on advances and future directions

of control methods in portfolio management, stochastic

models of markets and pricing, and hedging of options.

The ﬁnance area is highly interdisciplinary. This inter-

disciplinary nature comes from the necessity for a wide

variety of skills and knowledge bases.

As a major impetus for the development of ﬁnancial

management and economics, the research in ﬁnancial

engineering has had a major impact on the global econ-

omy. For instance, the Black–Scholes model [44.24]

and its various extensions for pricing of options [44.25]

using stochastic calculus have become a standard prac-

tice nowadays and have led to a revolution in the

ﬁnancial industry. Incidentally, it also won Scholes and

Merton the Nobel Prize for Economics in 1997 (Black

had unfortunately passed away 2years prior).

Powerful techniques of stochastic analysis and

stochastic control have been brought to almost all as-

pects of ﬁnance and resulted in a number of important

advances [44.12]. To name just a few, they include

the studies of valuation of contingent claims in com-

Part E 44.4

Education and Qualiﬁcation for Control and Automation 44.4 New Applications of Systems and Control Theory 771

plete and incomplete markets, consumption–investment

models with or without constraints, portfolio manage-

ment for institutional investors such as pension funds

and banks, and risk assessment and management using

ﬁnancial derivatives. At the same time, these appli-

cations require and stimulate many new and exciting

theoretical discoveries within the systems and control

ﬁeld. Take for instance the study of arbitrage theory,

risk assessment, and portfolio management, which have

collectively led to new developments in martingale the-

ory and stochastic control. Moreover, the development

of ﬁnancial engineering has created a large demand for

graduates at both Master and Ph.D. levels in industry,

resulting in the introduction of the curriculum in many

universities including Kent State University, Princeton

University, Columbia University, and the University of

California at Berkeley.

Another contribution to the control community from

ﬁnancial engineering and ﬁnancial mathematics was the

identiﬁcation and control of stochastic systems with

noise modeledby a fractional Brownian motion[44.11],

a process that can possess a long-range dependence.

Motivated by the need for this type of process in

telecommunications for models of Ethernet and asyn-

chronous transfer mode (AT M) trafﬁc, a new stochastic

calculus for fractional Brownian motion was developed.

This work in ﬁnancial engineering and ﬁnancial math-

ematics has since been successfully used in other ﬁelds

such as telecommunications and medicine, in particular

epilepsy analysis of brain waves [44.8,9].

44.4.2 Biomedical Models: Epilepsy Model

Epilepsy is a condition where a person has unprovoked

seizures at two or more separate times in her/his life.

A seizure is an abnormal electrical discharge within

the brain resulting in involuntary changes in move-

ment, sensation, perception, behavior, and/or level of

consciousness. It is estimated that 1% of the popula-

tions of industrialized countries have epilepsy whereas

5–10% of the populations of nonindustrialized coun-

tries have epilepsy. One link has been found between

epilepsy and malnutrition [44.26]. In the USA the num-

ber of epilepsy cases is signiﬁcantly larger than the

number of cases of people who have Parkinson’s dis-

ease, muscular dystrophy, multiple sclerosis, acquired

immunodeﬁciency syndrome (AIDS)orAlzheimer’s

disease [44.10]. The organizers of the Third Interna-

tional Workshop on Seizure Prediction in Epilepsy held

in Freiburg, Germany stated in the Welcome to the

Workshop that:

The great interest of participants from all over the

world and the highnumber of originalcontributions

presented ... instills conﬁdence in us that seizure

prediction is a promising ﬁeld for the years to come.

Epilepsy models, with their complexity, can serve

as an example of interdisciplinary and multidisciplinary

research for which systems and control approaches are

showing considerable promise. The methods pioneered

in the ﬁnancial mathematics and engineering sector de-

scribed above have been successfully used for detection

and prediction of epileptic seizures. There is published

evidencethat the seizure periodsof brain waves of some

patients have long-range dependencies with nonseizure

periods. Based on some initial work, it seems that the

estimates of the Hurst parameter, which is a charac-

terization of the long-range dependence in fractional

Brownian motion, have a noticeable change prior to and

during a seizure. Some of the algorithms [44.8]used

in identifying the Hurst parameter use stochastic calcu-

lus for fractional Brownian motion that was developed

within the ﬁnancial engineering and ﬁnancial mathe-

matics sector. A new application for the Hurst param-

eter, real-time event detection, has recently been identi-

ﬁed [44.9]. The high sensitivity to brain state changes,

ability to operate in real time, and small computational

requirements make Hurst parameter estimation well

suited for implementation into miniature implantable

devices for contingent delivery of antiseizure therapies.

This innovative interdisciplinary research has devel-

oped a new technology for future automated therapeutic

intervention devices to lessen, abort or prevent seizures,

opening the possibility of creating a brain pacemaker.

The goal is, by eliminating the unpredictability of

seizures, to minimize or prevent the disability caused

by epilepsy and hardship it imposes on patients and

their families and communities. It brings a hope to

improve productivity and better quality life for those

afﬂicted with epilepsy and their families as well as

for care-givers and healthcare providers. The creation

of a seizure warning device will minimize risks of in-

jury, the degrading experience associated with having

seizures in public, and the unpredictable disruption of

normal daily-life activities.

Only collaborative work of engineers, computer

scientists, physicists, physicians, biologists, and math-

ematicians can be successful in solving this type of

complex problem. Based partly on the success thus far,

there is a strong desire for a partnership with control

engineers. The University of Kansas (KU) Stochastic

Adaptive Control Group (SACG) has a long history

Part E 44.4

772 Part E Automation Management

of established collaboration in control education and

research with the KU Medical Center Comprehensive

Epilepsy Center and with Flint Hills Scientiﬁc, LLC

(FHS). FHS has one of the largest recorded collections

of long-term patient seizure data and it has used this

data to develop real-time seizure prediction algorithms

which have outperformed other prediction algorithms.

Recently FHS has initiated electrical stimulation con-

trol with a seizure prediction algorithm to prevent the

occurrence of seizures.

44.5 Pedagogical Approaches

The ﬁeld of control is currently suffering from a fun-

damental design ﬂaw. Because of a push from both

those external to thecontrol community (e.g.,the NAE’s

Engineer of 2020 reports) as well as those internal to

the control community, control has become an essen-

tial component for STEM education. This has resulted

in a ﬂood of “current-state” papers describing methods

and mechanisms that are currently being employed for

teaching control.

The beauty of control as an area of study lies in its

use of many different areas of mathematics: from func-

tional analysis through stochastic processes, stochastic

analysis, stochastic calculus of a fractional Brown-

ian motion, stochastic partial differential equations,

stochastic optimal control to methods of mathematical

statistics, as well as current computational methods in

stochastic differential and partial differential equations.

The curse of control as an area of study lies in its use

of many different areas of mathematics. Because of the

broad spectrum of mathematical tools for approaching

systems and control problems, it is not a subject that al-

lows for practical understanding without some amount

of deep coverage. Likewise, it is a topic that almost de-

mands some hands-on experimentation, which can be

both costly and time consuming. Several approaches are

described for building adequate and appropriate control

coursework into an already packed curriculum.

44.5.1 Coursework

A special pair of issues of the Control Systems Maga-

zine on innovations in undergraduate education [44.28,

29] presents a variety of articles related to undergradu-

ate controlsand systems education broken down intosix

broad categories; Kindergarten-12th grade (K-12) edu-

cation, course and curriculum development, experiment

development, special projects, software and laboratory

development, and lecture material.

Djaferis [44.27]describes a three-pronged approach

to introducing systems and controls in a ﬁrst-year engi-

neering course. A combination of lectures, simulations,

and a hands-on experiment where student develop a col-

lision avoidance system for a model car (Fig.44.3) are

used to indoctrinate students into how control theory

can be practically applied within the engineering design

process. Because the only prerequisite requirement for

the course is ﬁrst-semester calculus, it can be offered

during the spring term of the ﬁrst year, allowing stu-

dents theopportunity to experience controlearly in their

education. Additionally, the author found that the early

introduction to control resulted in a number of students

choosing to take a more advanced course in control as

well as for some to pursue graduate studies of systems

and control.

An example of a new introductory control course

for both engineers and general scientists is described

in [44.30]. The course was taught for over 10 years at

Sweden’s Lund University. The goals for the course

include:

•

Demonstrating the beneﬁts of control and the power

of feedback

•

Describing the role of control in the design process

and the importance of integrated systems

Fig. 44.3 Control laboratory experiment for ﬁrst-year en-

gineering students (after [44.27])

Part E 44.5

Education and Qualiﬁcation for Control and Automation 44.5 Pedagogical Approaches 773

•

Introducing the language of systems and control

•

Introducing the basic ideas and concepts of control

•

Explaining fundamental limitations

•

Explaining how to formulate, interpret, and test

speciﬁcations

•

Exploring computational tools such as MATLAB

and Simulink [44.32] to compute time and fre-

quency responses

•

Developing practical experience of sketching Bode

diagrams and performing calculations using the as-

sociated plots.

The challenge is to teach students sufﬁcient infor-

mation for understanding the validity of the information

contained in computer plots and to appreciate how to

achieve a satisfactory design that meets the speciﬁca-

tions and necessary criteria. One of the curriculardesign

ideas that fueled the course’s creation was to provide

a course that functioned as both an introduction for

those that elected to continue their control studies as

well as a functionally complete overview for students

who would only take one course on control.

One approach for demonstrating and appreciating

the importance of the dynamics of a system is the use of

the bicycle discussed in detail in [44.33] and proposed

for use in an introductory course. Because of the uni-

versal familiarity of the bicycle, it represents a highly

approachable problem context. It also presents a wide

variety of unique control- and dynamics-related prob-

lems; for example, the fact that when at rest it is an

unstable system but when in motion it exhibits aspects

of stability presents a unique problem situation for stu-

dents to understand. The problem can also be extended

into issues of rear-wheeled steering and other assorted

complexity inducing issues. Finally, the scope of the

problem is such that other non-control-related problems

can be introduced, such as discussions of the elemen-

Fig. 44.4 Autonomous vehicles competing in the DARPA Grand Challenge [44.31]

tal physics of motion and the theoretical aspects of

design.

In response to the National Academy of Engineer-

ing report [44.17,20] in which interdisciplinary system

engineering is cited as an increasingly important as-

pect of modern engineering and of the education of

future engineers projects that integrate elements of

mechanical design, modeling, control system design,

and software implementation were developed at the

University of Illinois at Urbana-Champaign [44.34].

In these projects, control design becomes an inte-

gral part of the larger systems engineering problems

and must be carried out in conjunction with design

and optimization of structural members, choice and

placement of sensors and actuators, electronics, power

considerations, modeling, simulations, identiﬁcation,

and software developments.

44.5.2 Laboratories as Interactive Learning

Environments

Understanding systems and control demands hands-

on experience. It is nearly impossible to appreciate

the complexities and interactions of a physical system

without witnessing it. Therefore a good control labo-

ratory is important for control education. Advances in

technology have reduced the cost of developing labora-

tories signiﬁcantly. Numerous courses currently utilize

the Lego Mindstorms kits which include a variety of

sensors and a programmable computer capable of in-

terpreting the sensor’s raw value [44.35–39]. While the

majority of those courses utilize the Lego system as

a vehicle for learning programming concepts or me-

chanical design principles, the sensor capabilities open

the door for use in a systems and control course.

The Intelligent Control Systems Laboratory [44.16]

at Australia’s Grifﬁth University demonstrates con-

Part E 44.5

774 Part E Automation Management

trol through cooperative driverless vehicles. The idea

of intelligent vehicles has brought with it promises

of heightened safety, reliability, and efﬁciency. In

2004, the Defense Advanced Research Projects Agency

(DARPA) held the ﬁrst of three DARPA Grand Chal-

lenge prize competitions for driverless vehicles [44.31]

(Fig.44.4). The ﬁrst year, the best vehicle drove only

7.36miles of the 150mile desert course. In 2006 at the

second challenge, ﬁve vehicles completed the course

and all but one of the 23 participating teams achieved

a distance greater than the maximum 7.36miles ob-

tained during the ﬁrst challenge. In 2007, the third

challenge required participants to navigate a more ur-

ban setting. Six of the 11 participating teams rose to the

challenge and successfully drove through the 60mile

urban course.

Finally, thenumber of interactive tools foreducation

in automatic control is growing rapidly. Many examples

are provided [44.40–43]. A sample plain talk presented

below provides other examples of creative, interesting,

and important modern control engineering education

laboratories.

44.5.3 Plain Talk on Control

for a Wide Range of the Public

The IEEE Control Systems Society Technical Commit-

tee on Control Education has been in the process of

developing a series of short presentations prepared for

a wide range of the public demonstrating the power,

beauty, and excitement of systems and control. These

presentations were given at the workshops for high-

school teachers and students, sponsored by the National

Science Foundation and the Control Systems Society.

They are presented in [44.44,45].

The following is a sample of such talks that were

presented at the 2007 Control and Decision Conference

in New Orleans:

•

T. Djaferis: The Power of Feedback (University of

Massachusetts Amherst)

•

C.G. Cassandras: Joys and Perils of Automation

(Dept. of Manufacturing Engineeringand Center for

Information and Systems Engineering Boston Uni-

versity)

•

R.M. Murray: Control Education and the DARPA

Grand Challenge (Control and Dynamical Systems,

California Institute of Technology)

•

P.R. Kumar: The Next Phase of the Informa-

tion Technology Revolution (University of Illinois,

Urbana-Champaign)

•

C. Tomlin: Controlling Air Trafﬁc (University of

California, Berkeley)

•

M. Spong: Control in Mechatronics and Robotics

(University of Illinois, Urbana-Champaign)

•

W.S. Levine, D. Hristu-Varsakelis: Some Uses for

Computer-AidedControl System Design Software in

Control Education (University of Maryland)

•

I. Osorio: Application of Control Theory to the

Problem of Epilepsy (University of Kansas Medical

Center and Mark Frei Flint Hills Scientiﬁc, LLC)

•

D. Duncan, T. Duncan, B. Pasik-Duncan: Random

Walk around Some Problems in Stochastic Systems

and Control (Yale University, University of Kansas)

•

K. Furuta: Understanding Phenomena through Real

Physical Objects–Controlling Pendulum (Tokyo

Denki University, Japan)

•

J. Baillieul: Risk Engineering–Past Successes and

Future Challenges (Intelligent Mechatronics Labo-

ratory, Boston University)

Developing plain talks that can be used by members

of the control community for noncontrol communities

in different settings is very important and has become

a major goal for the control education committees.

It is through these talks that the general population,

particularly children and young adults, can learn to ap-

preciate control and potentially pursue further study of

the topic.

44.5.4 New Approaches to Cultivating

Students Interest in Math, Science,

Engineering, and Technology

at K-12 Level

The IEEE CSS Technical Committee on Control Edu-

cation together with the American Automatic Control

Council (AACC) and the International Federation of

Automatic Control (IFAC) Committees on Control Ed-

ucation has organized a series of workshops for middle-

and high-school students and teachers on The Power,

Beauty, and Excitement of Control at all major control

conferences sponsored by CSS, AACC,andIFAC in

the USA and around the world since 2000 [44.46]. The

model for these workshops was created and developed

at the University of Kansas (KU) 15years ago. The uni-

versity organized semiannual half-day workshops for

ﬁfth- and sixth-graders from local schools to promote

STEM disciplines. They became very successful and

played a major role in establishingan importantpartner-

ship between K-12 and KU [44.47]. Another important

outreach activity for encouraging young people to con-

Part E 44.5