International Federation

1

• International Federation
• of Automatic Control
• EMERGING AREAS PROJECT
• Rotterdam August, 2003
• The goal of the on-going Emerging Areas project
  is to identify recent significant trends that are
  likely to continue within the automatic control
  field. The 2003 outcome of the project was a
  Panel Session. Presentations from that Session
  follow,
• Introduction to Emerging Areas Project
  Slide 7
• Integrated / Embedded Control
  Slide 17
• Distributed Control (over Communication
  Networks) Slide 24
• Collaborative Control
  Slide 30
• Hybrid/Discrete Event Systems/Networks
  Slide 36
• Autonomous Systems
  Slide 43
• Closing Comments Slide 48

2
IFAC NEWSLETTER

• IFAC Technical Board Identifies Emerging Areas
•  
• The goal of IFAC is to promote (in both theory
  and applications) science and technology of
  control in all systems, whether engineering,
  physical, biological, social or economic. In
  support of this goal, the IFAC Technical Board
  maintains on-going efforts to identify trends and
  forecast emerging areas within our field. The
  Technical Boards most recent formal activity
  regarding this thrust was a Workshop and Panel
  Session held in Rotterdam, The Netherlands, in
  conjunction with the IFAC Symposium on System
  Identification (SYSID).
•  
• The goal of the 2003 Emerging Areas Workshop /
  Panel was,
•  
•       Identify emerging trends within the
  control system and automation field
•       Forecast tomorrows most significant
  applications which will achieve higher
  performance, increased efficiency, lower cost, or
  other benefits
•       Identify the likely control methodologies
  and implementations that will enable such future
  improvements.
•  
• WORKSHOP Participants in the Workshop portion
  of the meeting were members of the IFAC Technical
  Board and selected Invited Guests from Dutch
  industries invited by the Dutch NMO. Each
  participant was asked to suggest
  trends/developments/needs within their respective
  areas of competence. As a result, over 50
  suggestions were presented and considered during
  the afternoon Workshop. (A complete inventory of
  all these suggestions is available at the IFAC
  web-site as described below.)
•  

3

•  
• After presentation and discussion of the various
  suggestions by each participant, several of the
  related suggestions were combined, and a final
  list was then prioritized to identify the
  following major trends believed to be especially
  significant
•         Increased Development of Theoretical
  Techniques and Practical Application of Hybrid /
  Discrete Event Systems
•         Increased Development and use of
  Distributed Control Techniques (Especially for
  Applications using Wireless communication
  Technology)
•         Continued Theoretical Development and
  Applications for Nonlinear Control to Overcome
  Limitations with Linear Representations / Models
•         Increased use and Development of
  Innovative, Ubiquitous Sensors and Actuators
  MEMS will Enable many New Sensors
•         More Effective Subsystems Operating at
  Optimal levels will be Integrated and Embedded to
  yield Improved Overall Systems
•         Interest in Learning Control Systems
  will Intensify the Name Adaptive Control may
  Not be Used, but the Concepts will Continue
•         Autonomous Robots and Autonomous Vehicle
  Development will Continue as Enabling
  Technologies Advance
•         Fault Detection / Isolation and
  Predictive Maintenance Techniques will greatly
  Improve operation of Highly Complex Systems
•         Collaborative Robots will assist and
  collaborate with Humans in Complex and/or
  Difficult Work Environments
•         Tools that Facilitate Collaborative
  Human-to-Human Work and Activities (e.g. E-Work
  and collaborative robotic systems) will Develop
•         New Hard and Soft Sensors will be
  Developed for Biotechnology and Biomedical
  Applications
•         Artificial Intelligence and Agent-Based
  Models will become more Useful for Controlling
  and Improving Economic Systems
•         Dramatic Automotive Control Advancements
  will Continue to Improve Safety, Operation, and
  Vehicle Efficiency
•         Autonomous Systems will Become Practical
  for Complex Operating Environments
•         Control Technologies will enable
  Realization and Deployment of next-generation
  high-performance Nano and Micro Systems future
  Controllers will benefit.
•  

4

•   
• PANEL SESSION The above trends were further
  reviewed and several were selected and refined
  for presentation in an evening Panel Session,
  attended by many of the SYSID Symposium attendees
  and members of the Dutch NMO. The allowable time
  for the Panel Session was of course limited, so
  only five trends were selected for the Panel
  Session. The highlights of these presentations
  are,
•  
• Integrated / Embedded Control
• Significant developments over the last decade
  within several different technologies now enable
  practical implementation of new control
  architectures. These advancements include
  innovative sensors actuators (many based on
  MEMS techniques), more powerful computers
  Digital Signal Processors, and exciting
  breakthroughs in communications network
  technology. As a result, two distinct trends are
  now gaining momentum within control. The first
  trend is integration such that perception
  (measurement, sensing) can now be embedded with
  control (controller, actuator) so that these
  previously separate functions are now
  transparent in fact, single components may
  someday sense, determine what to do, and then
  carry out the desired actions. The second trend
  is distributed control over wireless (and
  conventional wire-based) communication networks
  to connect embedded controllers into an overall
  closed loop operation. Such integrated/embedded
  controllers will enable improved medical
  technologies, increase energy efficiency, advance
  vehicle control, and enable many new consumer
  products.
•  
• Distributed Control (Over Communication Networks)
• Recent developments within both control theory
  and hardware tools now enable distributed
  control to become a practical reality. Numerous
  subsystems, each with their respective level of
  autonomy (but not co-located), can be integrated
  to control highly complex systems. As
  distributed control has progressed, it is common
  for large numbers of different types of
  components to exchange information through
  dispersed communication networks furthermore,
  recent developments of wireless communication
  offer yet another tool for this field. Although
  control, information theory, and communications
  are mature disciplines, theoretical issues in
  information theory and its effect upon
  performance of distributed control (conflict
  resolution, resource allocation, avoidance of
  deadlocks, etc.) are not well understood. Future
  control theory will address the impact of
  communication channel delays, quantization
  errors, transmission noise, random loss of
  information, as well as data handling and safety
  reliability. Attention will also be given to
  practical design of encoders, decoders,
  estimators, filters, and other communication
  elements to achieve improved performance,
  efficiency, and decisions made by large numbers
  of distributed controllers.
•  

5

•   
•  
• Collaborative Control
• Distributed systems are typically composed of
  numerous lower-level sub-systems with their
  individual control tasks and responsibilities.
  Collaboration among such interrelated systems is
  clearly essential in order to benefit from the
  respective strengths of the several partners.
  Fortunately, collaborative control trends are
  apparent for all system types. Machine-Machine
  Cooperation of smart robotic teams (including
  micro- and nano- as well as routine robots)
  will improve as new collaborative control
  techniques are developed as faults, errors, and
  interactions are better managed and as protocols
  for fault-tolerant operation are developed.
  Human-Machine Better understanding of how to
  share tasks will improve operation and will
  come from improved software, more incorporation
  of human factors, continued adaptation (learning)
  by the machine, better displays, new types of
  feedback, and new sensors actuators better
  tailored for human users. Human-Human Human
  team performance will also improve as
  enterprise software integrates and aids team
  decisions, as new methods (e.g.
  internet-conferencing) improve team coordination
  (even when remotely located with different
  databases, culture, or knowledge disciplines),
  and as task optimization methods enable multiple
  workers to share work and reduce local overloads.
  
• Hybrid / Discrete Event Systems / Networks
• Continuous (and discrete) Time systems and
  Discrete Event systems have essentially been
  developed independently of each other. However
  both phenomena frequently appear in the same
  process. Examples include manufacturing, process
  control (start-up/shut-down), autonomous, and
  distributed control systems. Radar processing
  provides a specific example which requires
  dynamic estimation of whether or not a target is
  present, what type target, whether or not it is
  maneuvering, forecasted track movements, etc.
  Typical solutions are sequential which involves
  detection, then classification, estimation, etc.
  Such approaches are clearly sub-optimal.
  Combined solutions for such Hybrid systems will
  no doubt develop in the future. New controllers
  will be driven by the need for higher performance
  from such systems, as well as the need to make
  better use of resources and an increased use of
  embedded/integrated systems. Theoretical
  developments are already being addressed and, as
  proven performance improves and reliability
  autonomy increase, the number of applications
  will grow. Controllers for such systems will
  probably be more complex than earlier solutions,
  and development of such methods will require
  merging of heretofore different fields and
  different designer approaches (sometimes even
  with different methods and solution languages).
•  

6

•   
•  
•  
• Autonomous Systems
• Todays automotive industry offers numerous
  innovations including driver assistance (ABS,
  ESP, Distance Detection), suspension control
  (passive), self-diagnostics, improved comfort
  (climate control, lighting, seats,
  entertainment), and more precise engine and
  driveline control. Emerging developments include
  drive-by-wire, brake-by-wire, parking assistance,
  collision warning, pedestrian detection, active
  suspension control, noise vibration control,
  and a host of telematics (navigation, on-board
  e-services, etc.). These developments will
  improve vehicle safety and, when coupled with
  infrastructure improvements, will yield
  intelligent traffic control. These developments
  will also leverage developments in autonomous
  unmanned vehicles for situations such as
  operation in hostile environments. Such vehicles
  will plan their own operations, as well as
  control the vehicle, to achieve these goals and
  develop alternate strategies when failures or
  unexpected hindrances are encountered. These
  autonomous concepts will also be extended to
  other applications such as unmanned factories and
  processing plants.
•  
• Post Session Comments 
• Various comments that have been submitted since
  the Panel Session begin at Slide 48 in this file.
  If you would like to add your comments, please
  send them to m.masten_at_ieee.org
• Mike Masten, Chair, Technical Board

7
IFAC TECHNICAL BOARD

• AUGUST, 2003
• IFAC EMERGING AREAS
• PROJECT
• INTRODUCTION TO PROJECT
• Presenter Mike Masten

8
IFAC EMERGING AREAS PANEL DISCUSSION

• Introduction to IFAC
• Goals
• Activities
• Goals of IFAC Emerging Areas Project
• Emerging Areas Process
• Panel Session
• Panelists Presentations
• Industrial Guest Evaluation Feedback
• Audience Comments Questions

9
INTERNATIONAL FEDERATION OF AUTOMATIC CONTROL

• Goal Promote science and technology of control
  in the broadest sense in all systems, whether,
  for example, engineering, physical, biological,
  social or economic, in both theory and
  application. Also concerned with the impact of
  control technology on society.
• Activities Organize technical meetings,
  publications, and any other activities consistent
  with IFAC constitution which enhances the
  interchange and circulation of information on
  automatic control activities.

10
IFAC MEETINGS

• CONGRESS
• Major IFAC event, Held every Three (3) years
• Recent Forthcoming Congresses
• 2002 Barcelona 2005 Prague 2008
  Seoul
• SYMPOSIA
• Long Term Events on Master Plan, usually Held
  Triennially
• CONFERENCES
• Technical Events, not Necessarily Part of a
  Series
• WORKSHOPS
• Smaller Events, less Formal

11
IFAC PUBLICATIONS

• IFAC NEWSLETTER
• Current IFAC News Information and Up-to-Date
  Announcements of Forthcoming Events
• AUTOMATICA
• Papers on Original Theoretical and Experimental
  Research and Development, Involving all facets of
  Control Theory and Applications
• CONTROL ENGINEERING PRACTICE
• Papers which Illustrate Applications of Control
  Theory and its Supporting Tools Emphasizes
  Practical Results
• ANNUAL REVIEWS IN CONTROL
• Best IFAC papers Presented at Meetings,
  re-written, and Broadened (or Commissioned
  Reviews in Emerging Research areas)
• JOURNAL OF PROCESS CONTROL
• Papers Relating to all Aspects of Chemical
  Process Control
• ENGINEERING APPLICATIONS OF ARTIFICIAL
  INTELLIGENCE
• Papers Relating to Intelligent Real-Time
  Automation
• AFFILIATED JOURNALS
• Usually associated with a Technical Committee
  List Available from IFAC Secretariat

12
IFAC TECHNICAL AREAS

13
IFAC EMERGING AREAS

• EMERGING AREAS PROJECT
• GOALS
• Identify the Major Emerging Trends within the
  Control System and Automation Field
• Forecast Tomorrows Most Significant Applications
  which will Achieve Higher Performance, Increased
  Efficiency, Lower Cost, or Other Benefits
• Identify Likely Control Methodologies and
  Implementations that will Enable Future
  Improvements

14
IFAC EMERGING AREAS

• PROCESS
• Workshop
• Consideration Potential Significant
  Trends/Forecasts
• Selection Most Significant Trends/Forecasts
• Preparation Breakouts to Prepare for Panel
• Panel Session
• Presentation Workshop Conclusions
• Evaluation Industrial Guest Feedback
• Discussion Audience Questions Comments

15
IFAC EMERGING AREAS

• PARTICIPANTS
• IFAC Technical Board Members
• Tohru Katayama Ruth Bars
  Robert Babuska
• Anibal Ollero Shimon Nof Denis Dochain
• Philipp Nenninger Keith Godfrey Talha
  Dinibutun
• Sirkka-Liisa Jamsa-Jounela
  Alberto Isidori
• Dongil Cho
• Industrial Guests
• Herman Van der Auweraer (Lueven Measurement
  Systems)
• Fred Abbink (National Aerospace Lab NLR)
• Ton Backx (IPCOS Technology)
• Hans Driessen (Thales)
• Alex van Delft (DSM)

16
IFAC EMERGING AREAS

• SOME OF THE CONSIDERATIONS
• Adaptive Control MEMS Driving Assistance
  Smart Drugs
• Robotics Manufacturing Artificial
  Intelligence Radar
• Optimization Navigation Speaking
  Animals Plants
• Telepresence Distributed Systems
  Wireless
• Software Agents Switching Control
  Learning Control
• Ubiquitous Sensors Actuators
  Perception Systems
• Stochastic Agriculture and
  Crops BioTechnology
• Economics Business Airports/Aerospace
  
• Transportation Systems E-Work Hybrid
  Systems
• Bio-Informatics PDE
  Systems
• Water/Waste Management Computers Control
• Infinite Dimensional Systems
• Intelligent Systems Modeling Identification

17
IFAC TECHNICAL BOARD

• AUGUST, 2003
• IFAC EMERGING AREAS
• PROJECT
• TREND INTEGRATED / EMBEDDED CONTROL
• Presenter Anibal Ollero

18
Panel Session PresentationEmbedded Control

• Ollero (Univ. Sevilla, Spain),
• R. Babuska (Delft University of Technology, The
  Netherlands) and
• H. Vander Auwerarer (IMS, Belgium)

19
     
EMBEDDED CONTROL
Embedded controllers
Control System
Embedded control separates control into
subsystems of the overall system, e.g. Hands,
Arms, Feet, Wheel
20
Enabling technologies

• Sensors and actuators with embedded
  intelligence.
• MEMS as supporting technology.
• Computer developments embedded systems with
  increasing powers.
• Communications/Networks.

21
Forecast/trends in control applications

• Medical technologies, health care surgical
  devices, medical instruments.
• Communication technologies network control,
  wireless devices.
• Advanced vehicle control.
• New cars, vehicles and transportation
  technologies.
• Technology for energy savings
• mixed energy sources, small scale energy
  systems.
• Consumer products, instruments, MEMS.

Performance, reliability, cost, easy to use and
maintain.
22
Most significant technological trends

• Two poles
• Integration
• Embedded control systems integrating perception
  and control functions that can be used,
  eventually, in a transparent way.
• Reasons Technological developments (Hardware
  integration, MEMS, ) and new applications
  (vehicles, autonomous systems, consumer products,
  biomedical systems, )
• Distribution
• Distributed control systems with wire and
  wireless connections between components and
  embedded controllers.
• Reasons Communications technology and new
  applications (Tele- applications, distributed
  manufacturing, protection of people and
  environment, home automation, .)

23
The single trend/forecast

• Integration of control and perception components
  in embedded systems that could be networked using
  wire or wireless technologies.

24
IFAC TECHNICAL BOARD

• AUGUST, 2003
• IFAC EMERGING AREAS
• PROJECT
• TREND DISTRIBUTED CONTROL (OVER COMMUNICATION
  NETWORKS)
• Presenter Philipp Nenninger

25
Control of systems distributed over communication
networks

• R.Bars, A.Isidori, P. Nenninger, A.Ollero

26
Motivation

• More and more systems are becoming distributed,
  consisting of a large number of components of a
  very different nature, which exchange information
  through wire/wireless networks (Wireless
  communications in the loop)
• While control, information theory communication
  are mature disciplines, little effort has been
  put so far in understanding how issues in
  information theory affects the performance of a
  distributed control system

27
Theoretical Challenges

• Design of
• Encoders
• Decoders
• Communication channels
• Controllers / estimators
• To achieve prescribed performances with minimum
  loss, high efficiency and with decisions made by
  a large number of users

28
Theoretical Challenges

• Quantitative analysis of how the performance of
  the system is affected by
• Bandwidth
• Delays
• Quantization errors
• Transmission noise, loss of information
• Data handling / control of data flow
• Safety / reliability issues
• Conflict resolution / avoidance of deadlocks
• Resource allocation

29
Emerging Application Areas

• Technologies for safety critical and hostile
  environments
• Remote control and coordination of unmanned
  vehicles (UAV, UGV, AUVs)
• Telepresence
• Remote laboratory
• Remote surgery
• Distributed manufacturing
• Home automation
• Ubiquitous sensors

30
IFAC TECHNICAL BOARD

• AUGUST, 2003
• IFAC EMERGING AREAS
• PROJECT
• TREND COLLABORATIVE CONTROL
• Presenter Shimon Nof

31
Trends in Collaboration Presented by Shimon Y.
Nof Purdue University, USA and Stephen
Kahne Embry Riddle University, USA For the IFAC
Technical Board Meeting Rotterdam, Holland,
August 2003
32
Machine Machine
Overall Trend Smart robotic teams (normal,
micro, nano robots) will be able to interact even
better than human teams Trend 1. Collaborative
Coordination Control Theory Safety is
critical Enable unmanned manufacturing and
maintenance Trend 2. Control Methods to Manage
faults, errors, conflicts, and interactions
Save money Improve product uniformity Trend
3. Control Protocols for fault-tolerant,
time-out integration of information
signals Future collaborative machines will
depend on cheaper, redundant arrays/networks
(e.g., FTTP)
33
Human -- Human
Overall Trend Smart tools and collaboration will
enable significantly better complex system
performance Trend 1. Interoperability of
software for enterprise applications
Integration of team decision support, e.g. air
traffic control, ERP, supply networks Trend 2.
Coordination of team members interactions
Geographic remoteness of collaborating members
Members have different databases, culture,
knowledge Trend 3. Optimization of parallelism
among resources Two (or more) heads are
better than one Overcome overload of tasks
34
Human -- Machine
Overall Trend Better understanding of how to
share tasks Trend 1. Impedance matching
Software Human factors -- information
usability Process goals adjustment-adaptation
(learning) Trend 2. Performance monitoring
Better displays New forms of feedback Trend
3. New sensors and actuators Fly by
everything e.g. , by wire , light
Understand who is the human customer/partner/user
35

• CONCLUSIONS
• Collaboration in distributed operational systems
  is critical
• Collaboration must be optimized to benefit from
  respective strengths of the partners
• Control theory and applications must be developed
  and verified to enable this new control
  environment trend

36
IFAC TECHNICAL BOARD

• AUGUST, 2003
• IFAC EMERGING AREAS
• PROJECT
• TREND HYBRID / DISCRETE EVENT SYSTEMS / NETWORKS
  
• Presenter Hans Driessen

37
IFAC Emerging Areas Panel Session
Presentation Hybrid/Discrete Event
Systems/Networks Hans Driessen Thales, The
Netherlands Pedro Albertos Universidad
Politecnica de Valencia, Spain                 
38
Hybrid/Discrete Event Systems/Networks
Motivation Continuous (and discrete) time
systems at one hand and discrete event systems on
the other hand have been studied / developed
independently of each other, but often appear in
the same process. An incomplete list of
application areas isdistributed control
systems, autonomous systems, manufacturing,
process control (start-up/shut down), radar
processing management. 1. What is the
forecast? The forecast is that we will see a
combined treatment in the future.2. Why is this
forecast likely to happen? This is due to the
ever higher performance requirements, e.g. a
shorter dead-time. Also better use of the
resources, Embedded systems, and Integrated
control design.
39
Hybrid/Discrete Event Systems/Networks
3. When is this forecast likely to
happen? Theoretical developments are taking
place right now.

• Theoretical Challenges
• Integrate the control design and its
  implementation
• Develop the joint theory for hybrid systems and
  their control
• Include the delays and randomness of the
  communication networks

Applications will appear in coming years.
40
Hybrid/Discrete Event Systems/Networks
4. What will be the impact when this
happens? Higher performance of systems. Broader
spectrum of applications. Increased reliability
and autonomy.5. Are there any adverse effects
of this development? Increased complexity of the
problem /solutions6. What must happen for this
forecast to come true? Combine different fields,
people, even language- Control and Real Time SW
expertise (difficult)- Continuous/discrete time
control and discrete event systems (easier to
combine)
41
Hybrid/Discrete Event Systems/Networks
IllustrationsSome of these ideas can be
illustrated with a typical area of applications
namely, radar systems.  In radar processing we
encounter dynamic estimation problems including
both discrete and continuous state variables. The
discrete variables amount to whether an object is
present or not, whether it is maneuvering or not,
what type of object it is from a known class of
possible objects. So infact there are
simultaneous, dynamic, stochasticdetection/estima
tion/classification problems. In radar
management we encounter both continuous and
discrete parameters to be selected on-line for
every transmission in order to maximize
performance. This involves calculations of
expected performance for hybrid systems and
control schemes for obtaining the maximum of
performance.
42
Hybrid/Discrete Event Systems/Networks
Illustrations ContinuedUp to
now most of the simultaneous detection /
estimation / classification problems are treated
sequentially, more or less independent so we
first try to detect, then estimate, and then
classify. This is sub-optimal and leads to all
kind of smart and tricky algorithms. This is not
only sub-optimal, but also hard to reuse from
application-to- application. Significant
performance improvements can be obtained by
solving the problems simultaneously, but this
induces a computational problem that cannot be
solved with traditional techniques, or at least
very difficult. Particle filtering, or
Sequential Monte Carlo filtering, is a technique
that is very well-suited for solving problems
including discrete variables, nonlinearities,
constraints etc. These algorithms lead to
well-understood algorithmic solutions that are
easily re-used, saving considerable development
time and costs.
43
IFAC TECHNICAL BOARD

• AUGUST, 2003
• IFAC EMERGING AREAS
• PROJECT
• TREND AUTONOMOUS SYSTEMS
• Presenter Anibal Ollero

44
Panel Session PresentationAutonomous Systems

• Ollero (Univ. Sevilla, Spain),
• R. Babuska (Delft University of Technology, The
  Netherlands) and
• H. Vander Auwerarer (IMS, Belgium)

45
Industry Trend Advanced Vehicle Control

• Increasing role of vehicle electronics.
• Current functionality
• Engine control
• Driver assistance (ABS, ESP, distance detection)
• Suspension control (mainly semi-active)
• Diagnostics (ABS)
• Comfort (climate control, lighting, seats)
• Supported by standards such as CAN bus etc.

46
Industry Trend Advanced Vehicle Control
Emerging functionality X-by-wire
drive-by-wire, brake-by-wire. Driver assistance
(parking assistance, speed control, dangerous
maneuvering) Active safety measures (collision
warning, pedestrian detection) Active suspension
control Active noise and vibration
control Telematics (on-board e-services,
navigation and beyond)
47
Autonomous systems

• Autonomous vehicles
• UAV, UGV, AUVs, multiple vehicles
• Autonomous functions in conventional vehicles
  (cars, aircraft, vessels, ..)
• Driver overrule (collision avoidance, lane
  control, ..)
• Intelligent Traffic Control
• Technologies for safety-critical and hostile
  environments
• space, disaster remediation, defense, ..
• Combination of autonomy and Teleoperation.
• Unmanned plants.
• Application of learning, uncertainty handling and
  AI techniques.
• Autonomous perception.
• Reactivity and planning techniques.
• Reliability is a main issue.

48
IFAC TECHNICAL BOARD

• AUGUST, 2003
• IFAC EMERGING AREAS
• PROJECT
• CLOSING COMMENTS

49

• Closing Comments
• The comments on the following slides were
  received after the Workshop and Panel Discussion.
  These comments do not necessarily represent the
  collective opinions of the participants, but
  rather the individuals who submitted them. These
  comments are provided to stimulate further
  discussion of this important subject.
• If you would like to contribute additional
  comments, please sent them to Mike Masten at
  m.masten_at_ieee.org Comments regarding trends that
  you believe will be highly significant within our
  field are especially welcome. We will
  periodically update this section of this
  presentation if such comments are submitted.

50

• I wrote down some remarks/items on which I will
  be happy to contribute more.
• The summary discussion highlighted some of the
  developments (in theory, technology and
  applications) that are linked to our field of
  research. It appeared to me that the audience
  liked the structured way in which Shimon Nof
  presented the different trends M-M, H-H, H-M. It
  was hard to find trends (in his and other
  presentations) that did not imply 'more of the
  same or make things better' (in terms of
  reliability, working domain, robustness).
• It should be stated that complexity is the enemy
  of reliability. Hence, understanding complexity
  and trying to cope with it are key issues in
  theory and technology developments and in the
  design of applications. Trends that host 'the
  complexity virus' are the drive to increase
  automation, implement intelligent (autonomous)
  sensor- and actuator systems, link/teleconnect
  systems that are different in nature (causing all
  kinds of unexpected 'dynamics') and that are at a
  far distance.
• I personally disagree that the gap between
  practical problems and research at the university
  is so big. I think industry is not capable to
  formulate their problems such that these can be
  worked on by academia and serve as an educational
  podium. Industry and academia have different
  missions!
• I hope these remarks are helpful for further
  discussion.
• _______________________
• Peter A. Wieringa Man-Machine Systems,
  Mechanical Engineering
• Faculty of Design Engineering, Delft
  University of Technology TU Delft

51

• I am uncomfortable with Peter Wieringa's
  comments. He more or less reduces the sense
  of urgency to close the gap between academia and
  industrial practice (the well known time-delay),
  whereas in the Panel Discussion we were quite in
  agreement about the gap and the need to overcome
  it. Remarks made at the end of the session about
  the thoroughness of the meeting clearly suggested
  that we should repeat these kind of sessions and
  to devote more time to this challenge.
• Anonymous Comment
• I agree the gap between practical problems and
  university research is not that large, but we
  should not try to identify who is responsible for
  the gap.  I believe we have to avoid a
  formulation where it may appear that the problem
  is in "the other camp". The most fruitful
  approach, which was also the spirit of the panel
  meeting, is "What we can learn from each other",
  industry by (early) understanding the trends in
  fundamental research (what is possible), academia
  by understanding the (long-term) industrial needs
  and applications (what is needed, where can this
  be actually used). While the missions are indeed
  different, in the long run, they touch. We both,
  industry and academia, have to look beyond the
  scope of our classical, daily, way of thinking,
  and this requires efforts from both sides. The
  result is a cross-fertilization of ideas, methods
  and applications that does not follow a simple
  linear model (such as industry needs -gt
  university develops -gt industry uses). I am
  convinced many good examples of such efforts
  exist and I felt that the panel meeting was held
  in this spirit. 
• Herman Van der Auweraer

52

• I do not fully agree with the remark of Peter
  Wieringa saying that industry is not capable to
  formulate their problems such that these can be
  worked on by academia ... etc. I accept it as a
  personal remark and opinion, but we should
  perhaps be a little careful that IFAC is not
  identified with this (slightly arrogant) view.
• Anonymous

53
IFAC EMERGING AREAS

• THANK YOU
• To IFAC Technical Committee and Coordinating
  Committee Chairs
• Identification of Potential Forecasts/Trends
• Presentation of Forecasts/Trends for
  Consideration
• Participation in Workshop and Panel Session
• To Industrial Guests
• Identification Presentation of Potential
  Forecasts/Trends
• Participation in Workshop Panel Session
• Evaluation/Feedback at End of Panel Session