CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.

XXI - Human-Machine Interaction - Gunnar Johannsen

HUMAN-MACHINE INTERACTION

Gunnar Johannsen
Department of Mechanical Engineering, University of Kassel, Germany

Keywords: auditory display, control device, decision support system, dialogue system,
fault management, functional display, human-centered design, human-computer
interaction, human error, human-machine interface, human-machine system, human
operator, knowledge-based system, multimedia, process visualization, systems life-
cycle, task analysis, usability engineering, user participation, visual display.

Contents

1. Introduction

S
TE S
2. Human Tasks with Automation and Control
3. Human-Machine Interfaces

R
AP LS
3.1. Traditional and Advanced Functionalities
3.2. Dialogue Systems
3.3. Control Devices
C EO
3.4. Visual Displays
3.5. Functional Displays and Information Preprocessing
3.6. Auditory Displays
3.7. Multimedia and Multimodal Displays
E –

3.8. Adaptive Interfaces

H

4. Knowledge-Based Support
PL O

4.1. User-Oriented and Application-Oriented Functionalities

4.2. Procedural Support
M SC

4.3. Human Error Prevention and Recovery

4.4. Fault Management Support
5. Design and Evaluation
6. Conclusions
SA NE

Glossary
Bibliography
Biographical Sketch
U

Summary

Human-machine interaction is described as the interaction and communication between

human users and a machine, a dynamic technical system, via a human-machine
interface. The real-time aspect distinguishes the fields of human-machine systems and
human-computer interaction which are otherwise strongly related. Research in human-
machine interaction has been performed for more than 50 years. Good designs of
human-machine interaction and systems have gained a high market value for many
products and services in application domains such as industrial, transportation, medical,
and entertainment systems.

The main human task categories in human-machine interaction are controlling and
problem solving. Controlling comprises continuous and discrete tasks of open- and

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

closed-loop activities whereas problem solving includes the higher cognitive tasks of
fault management and planning.

In this chapter human-machine interfaces are introduced with traditional and advanced
functionalities. Presentation, dialogue, and control input are the more traditional levels.

Advanced functionalities provide explanation and adaptability based on user and

application models. Dialogue systems and control devices are explained. The
presentation issues are elaborated with different types of displays.

The visual displays use computer graphics for process visualization through topological,
component-oriented views with dynamic pictures. Complementary views can supply
functional knowledge. Auditory displays are becoming more important, with indicating
alarms and state information, also as a means for avoiding visual overload. The
integration into multimedia leads to multimodal displays.

S
TE S
R
AP LS
Knowledge-based systems for decision support are another advanced means for
improving human-machine interaction. Procedural support and human error prevention
are examples for user-oriented functionalities.
C EO
Fault management support, particularly the diagnosis of technical failures, exemplifies
application-oriented functionalities. The design and the evaluation of human-machine
systems components are outlined with task orientation and different forms of user
participation.
E –
H

1. Introduction
PL O

Human-machine interaction with industrial plants and other dynamic technical systems
M SC

has nowadays been recognized as essential for process safety, quality, and efficiency. It
comprises all aspects of interaction and communication between human users and their
machines via human-machine interfaces.
SA NE

The whole system of human users, the human-machine interface, and the machine is the
so-called human-machine system (HMS); see Figure 1. Different human user classes
U

may be involved, namely, operators, engineers, maintenance personnel, and managers.

They have different but overlapping information needs.

The term “machine” indicates any kind of dynamic technical system (or real-time
application), including its automation and decision support equipment and software –
and it relates to many diverse application domains. The automation components of the
technical system are denoted as supervision and control systems.

They interact directly with the pure technical (production) process. Examples of such
processes are a power generation process, a chemical or a discrete-parts production
process, an aircraft, a tele-manipulator or a real-time software application.

The decision support systems are more advanced, knowledge-based functionalities of

the machine which provide advice for the human users.

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

Figure 1: The whole human-machine system (HMS) with human users, human-machine
interface, and the machine.

The application domains include all kinds of industrial, transportation, medical, service,
home, and entertainment systems. The more traditional application domains of
industrial and transportation systems have nowadays been supplemented by application
domains from medical to entertainment systems. All application domains profit from
advancements in the development of the human-computer interaction field. Human-

S
TE S
computer interaction (HCI) is characterized as human interaction with those computer

R
application domains which are not determined by a dynamic process or by real-time

AP LS
constraints, as it is always the case in human-machine interaction (HMI).

The degree of automation in control of dynamic technical systems has substantially

C EO
been increased over the last decades. This is true for all technical systems such as power
plants, industrial production plants, and vehicles and transportation systems. High levels
of safety, performance, and efficiency have been achieved by means of the increased
use of automatic control. Interestingly enough − however, with no surprise to human
E –

factors specialists − the need for improved human-machine communication increased

H

(rather than decreased) with the increased degree of automation. Generally, increased
PL O

automation does not replace the human users who are interacting with the machine, but
shifts the location of the interface between both. The machine as an automated system
M SC

becomes more complex than one with less automation, leading to more sophisticated
structures of supervisory control. Higher complexity and more sophisticated control
structures require a new quality of communication and co-operation between human and
SA NE

machine. The role of the human user shifts from that of a controller to that of a
supervisor. The human supervisor interacts with the process through one or several
layers of computers on which the human-machine interface, the automation, and the
U

decision support functionalities are implemented.

Thus, the functionalities of well-designed human-machine interaction have gained

predominant importance in advanced process control as a consequence of the degree of
automation and, also, of the increase of extended automation functionalities. The
advanced technologies with extended functionalities include expert systems for control
and diagnostics, fuzzy process control, and artificial neural networks in process control.
It is important to ensure integration between conventional control systems and these
advanced technologies for the purpose of reaching the overall goals of efficient and safe
systems performance as well as sustained and satisfying human performance.

Matching the efforts of automation and of human factors issues leads to the approach of
human-centered automation, recommended by many researchers for quite a long time.
Today, it is also more and more followed in industry. Under the perspective of human-

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

centered automation and human-centered design, human-machine interaction appears to

be a symbiotic process between the human users and their machines via human-machine
interfaces whereby both − machines and interfaces − have properly been designed. The
functionalities of both, the machine and the interface, have to be specified and designed
with a strong view on human-centeredness, i.e., with respect to user orientation −
concerning human cognitive processes, human needs, and human capabilities − and
based on goal orientation and task orientation. Then, these functionalities can be im-
plemented as extended automation modules and knowledge-based decision support
systems within the machine and as interaction and communication modules within the
human-machine interfaces. Computer graphics, multimedia and multimodal displays, as
well as knowledge-based and other software technologies offer a wide range of
alternative designs for human-machine interfaces, extended automation, and
knowledge-based support.

Human-machine interaction is goal-oriented. The overall goals of human-machine

S
TE S
systems are mainly (1) productivity goals, (2) safety goals, (3) humanization goals, and

R
AP LS
(4) environmental compatibility goals. The productivity goals include economic as well
as product and production quality goals. The importance of the safety goals is strongly
influenced by the application domain. This goal class dominates all others in many
C EO
large-scale systems and, particularly, in risky systems. The humanization goals
comprise team and work organization, job satisfaction, ergonomic compatibility, and
cognitive compatibility. The latter includes the sub-goals of transparency and human
understanding. The environmental compatibility goals refer to the consumption of
E –

energy and material resources as well as to impacts on soil, water, and air.
H

Human-machine interaction and human-machine systems research require multi-

PL O

disciplinary or interdisciplinary views and approaches. The following three domains

M SC

contribute to human-machine interaction and systems research: (1) cognitive science

and ergonomics (as the human sciences), (2) automation and systems engineering (as
the systems sciences), and (3) information and communication engineering (as the
computer sciences). Further, organizational and cultural aspects are strongly involved.
SA NE

Human-machine systems research has now been performed for about 60 years.
Nevertheless, it took quite some time before the human-machine systems field became
U

an established and more widely accepted discipline, particularly also in industries. For
long time, mainly the aeronautics and astronautics industries and, to a lesser extent, the
car manufacturing industries showed some more interest. This situation has now totally
changed. It has been recognized that a good symbiosis between human and machine is
generally required. Good human-machine interaction and systems designs have gained a
high market value for almost all products and services. Consequently, strong needs in
industry and society exist today as an important application pull. This influences the
human-machine systems field which became mature but continues, at the same time, as
an innovative and always further developing research area.

2. Human Tasks with Automation and Control

It is important to consider cognitive capabilities and information processing behavior of

human users interacting with the technical systems, as well as the tasks to be performed

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

within any human-machine system, before and beyond all technological possibilities.
Task orientation and human-centered designs are necessary for improved operational
use of technical processes and their human-machine interaction.

Any user-oriented design of interactive products, such as human-machine systems,

should start with definitions of user classes, overall goals and means for their
achievement, as well as with task analyses, in order to have a solid basis for user
requirements and systems specifications, particularly also for the functional specifi-
cations of the human-machine systems components and functionalities. User
requirements should consider the goals-means-tasks relationships, and should be based
on a task-oriented perspective.

Two principal task categories can comprise all human activities in human-machine
systems, namely controlling and problem solving. All other tasks such as monitoring
and communicating can be classified as subtasks or supporting tasks. Figure 2 shows a

S
TE S
schematic block diagram of a conceptual human performance model together with the

R
AP LS
automated technical system with their interactions via control devices and displays for
controlling and problem solving with respect to prescribed goals which have to be
matched by the system’s outputs. Controlling comprises continuous and discrete tasks
C EO
of open-loop and closed-loop activities. The higher cognitive tasks of problem solving
include fault management (with fault detection, diagnosis, compensation, and
correction) as well as planning. The two task categories require a different selection of
information about goals and systems outputs. While the perception of deviations
between goals and outputs is normally sufficient for controlling, both types of inputs
E –

have to be interpreted more consciously for problem solving tasks. Further, the problem
H

solving tasks are much more knowledge-based activities than the controlling tasks.
PL O
M SC
SA NE
U

Figure 2: Human performance model with controlling and problem solving, interacting
with the automated technical system (after Johannsen, 1992).

Roughly speaking, the two human task categories of controlling and problem solving
correspond to the two categories of computer use in advanced automated technical

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

systems, namely computer control (by supervision and control systems) and computer
support (by decision support systems); see also Figure 3. This includes the more
traditional levels of automation systems as well as all levels of knowledge-based
systems. Only since the latter became a more mature technology over the last years, the
possibilities of function and task allocations between human operators and computers
allow us to relate both human task categories to corresponding categories of computer
use. That means also that truly human-centered designs of human-machine systems can
now better be pursued and implemented with the latest technology of knowledge
support.

The interactions between supervision and control systems, decision support systems,
and human operators have to be carried out in such a way that goals and their subgoals
are achieved with respect to the technical process and the overall system. The goals and
appropriate goal structures are either prescribed or derived by the human operators.
Thus, some of the major responsibilities of human operators are the selection of goals

S
TE S
and subgoals, the formation of goal structures, and the checking of goal accomplishment

R
AP LS
and subgoal compatibility.

A closer look at the two task categories of controlling and problem solving shows
C EO
similarities and differences between their human form and its automated counterpart.
The behavior of human operators and, in principle, also of automation systems can be
classified into the three information processing phases of categorization, planning, and
action. These three phases have to be accomplished in all controlling and problem
solving tasks, either on a state-oriented, context-oriented or structure-oriented level.
E –

Here, context refers to the fields of information involved. The three levels correspond
H

with the cognitive levels of skill-based, rule-based, and knowledge-based human

PL O

behavior.
M SC

Human operators of technical systems can much more flexibly maneuver through all
these different possibilities than the automation systems. However, they may be under
time pressure or otherwise impaired or overloaded in certain, particularly critical
SA NE

situations. Therefore, the state of computerization of the technical systems has often
been developed quite far. This again has sometimes led to the so-called “ironies of
automation”. For example, boredom may occur in normal situations with highly
U

automated systems, and boredom may lead to loss of vigilance, poor job satisfaction,
and possibly human errors. Human-centered designs of automated technical systems
have to avoid these effects by choosing a flexible function and task allocation between
human operators, automation systems for supervision and control, and knowledge-based
systems for decision support under all normal, abnormal, and emergency conditions.

The issues of human-machine interaction are not only concerned with the interaction
between a single human and the dynamic technical system, the machine. Real work
situations in many application domains require that several humans of different human
user classes frequently interact with the same technical system and communicate with
each other. Some examples of such technical systems are chemical plants, cement
plants, power plants, transportation systems, and discrete manufacturing systems.

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

In plant-wide control and communication, several partially overlapping groups exist.

Their official hierarchical structure is often complemented by an informal heterarchical
or even flexible opportunistic organization, dependent on company cultures, social and
personal preferences, and specific work requirements. As an example, the application
domain of the cement industry shows that several persons from different human user
classes need to co-operate, in the plant control room or, also, in office rooms during
group meetings. These people from different human user classes share their overlapping
skills and knowledge, and interact with each other, at least partially, via the human-ma-
chine interfaces or directly face-to-face, in addition to communicating with the technical
system also via the human-machine interface.

3. Human-Machine Interfaces

3.1. Traditional and Advanced Functionalities

S
TE S
The human-machine interface provides the information links between one or several

R
AP LS
human users and the machine (i.e., the technical system or application). This is shown in
Figure 3. Therein, the four main classes of human users are exemplified, namely
operators, engineers, maintenance personnel, and managers. The machine consists of its
C EO
technical process, as well as its supervision and control systems for automation, its
decision support systems for knowledge-based advice, and its data and knowledge
libraries.
E –
H
PL O
M SC
SA NE
U

Figure 3: Main functionalities of the human-machine interface between human users

and the machine (after Johannsen, 1997b; and Johannsen, Heuer, and Tiemann, 1997).

As also outlined in Figure 3, the human-machine interface is subdivided into

components for presentation and control (as the human interface), dialogue and
information preprocessing, and the application interface as the more traditional
functionalities, and into user model, explanation and tutoring, and application model
components as the more advanced functionalities. The latter include also new
approaches towards improved information preprocessing.

The more traditional levels of human-machine interfaces, such as presentation and

dialogue, are generally separable from each other. The presentation and control level is

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

concerned with the problems of how to present the information to the human users, and
how to transform their control inputs to the machine. The dialogue level deals with the
information flows regarding such problems as what information to handle when, i.e., the
handling of the adequate information at the appropriate time, between the human users
and all other components of the whole human-machine system. Information
preprocessing is more and more often applied in order to provide improved information
context and, hence, to facilitate the information processing activities of the human users.

The presentation and control level can also be viewed as the human interface in the
narrower sense, on the human side. It is contrasted by the pure technological application
interface on the machine side. Generally, visual, auditory (non-speech sound and
speech), gestural, mimic (including face-to-face), and haptic (sense of touch) informa-
tion as well as vibrations can be used as forms of display for presentation.

Combinations lead to multimedia and multimodal displays which will become more

S
TE S
important in the future. Currently however, the main mode of presentation is still the

R
AP LS
visualization, and the graphical user interface is and will be, for some time,
predominant. In former times, visual displays were implemented with electromechanical
instruments. Today, they are replaced by computer graphics systems which contain a lot
C EO
of graphical and textual dynamic pictures. These pictures are created with dynamic
graphical editors.

Functional displays have been derived from the more traditional visual displays. They
include new information preprocessing modules for a better consideration of user and
E –

goal orientations. Thereby, these displays realize more advanced functionalities.

H
PL O

Both, the presentation and the dialogue levels can explicitly depend in their
functionalities on the goals of the system as well as on knowledge-based technical
M SC

systems or application models, user or operator models, and task models. The more
explicit representation of such models in the human-machine interface leads to more
advanced paradigms. An application model contains the knowledge about the goals, the
SA NE

structure and the functions of a particular application. The functionality of this

application or technical systems model internally supports all the other functionalities of
the human-machine interface.
U

A user model functionality is needed if a certain adaptability to human user classes or

single users shall be achieved. This is the major ingredient for adaptive interfaces. A
more elaborated user model will always include a technical systems model in order to
represent the user's view with respect to the technical system. In addition, knowledge on
human information processing behavior and cognitive strategies has to be represented in
a user model by means of algorithms, rules, and, possibly, active inference mechanisms.

Another knowledge-based functionality of a human-machine interface is the explanation

functionality. It is a kind of knowledge-based online help system which informs the
human users on request about what the components of the technical system (and
possibly also of the human-machine interface) mean and how they function. A tutoring
functionality supplies even more knowledge in an interactive way to novices and
occasional users.

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

3.2. Dialogue Systems

The dialogue functionality of Figure 3 can be designed as a quite complicated goal-

oriented component for handling the dialogue between the human users and all other
components and subcomponents of the human-machine interface and of the technical
system. Among other subfunctionalities, this includes the resolution of possible
conflicts between different components with respect to dialogue requests to and from
the human users.

Different paradigms exist for the dialogue functionality in human-machine interfaces.

Three possible roles of the dialogue functionality are differentiated here. With the first
paradigm, the dialogue system is a kind of knowledge-based intermediary between the
presentation system and both the supervision and control system as well as the decision
support system as parts of the technical system. Below, this paradigm will be explained
a little further.

S
TE S
R
AP LS
The second paradigm of the dialogue functionality exemplifies the dialogue as a
communication language or mechanism between the human users and all the
subsystems of the technical system. This kind of implicit view on the dialogue in
C EO
human-machine communications can often be found in the literature. Direct manipula-
tion and menu selection are communication mechanisms which are suitable for
dialogues between human users and dynamic technical systems. The third paradigm
understands the dialogue system as a co-operative assistant to the presentation system.
In this case, the dialogue system is a kind of controller or manager for the information
E –

flows through the presentation system.

H
PL O
M SC
SA NE
U

Figure 4: Paradigm of the dialogue system as a knowledge-based intermediary between

presentation system and technical system (after Alty and Johannsen, 1989).

An example for the first paradigm is shown in some more detail in Figure 4. A module
for handling dialogue events and dialogue event sequences connects all the subsystems
of the technical system with the presentation system. Urgent requests, such as alarm
messages, from some subsystems, as well as urgent requests from the human users, may
interrupt currently ongoing dialogues with other subsystems. Then, the dialogue channel
is correspondingly switched by the handler for interrupts and scheduling. Thereby, loose
ends of non-terminated (because interrupted) dialogues may occur. By means of a

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

dialogue knowledge base, the human user can be supported in resuming such loose ends
of dialogue. Dedicated dialogue assistants aid with time- and contents-related knowl-
edge about their respective technical subsystems. The time aspect and the question of
how much knowledge of the subsystems needs to be held locally by the dialogue
assistants, depend heavily on the task situations.

3.3. Control Devices

A very broad spectrum of control functionalities as possibilities of control input in

human-machine interaction exists. Different forms of control can be exerted, either
direct (e.g., gestural as with the conductor of an orchestra), via bio-signals or via control
devices. Control via bio-signals is of particular value for handicapped people, e.g., in
the operation of an arm prosthesis. Speech and motion (displacements, forces, and
gestures) can be used directly or via control devices. Speech input and gestural control
will eventually become new means of control in the future.

S
TE S
R
AP LS
A few operational systems are already in use. Speech recognition allows the automatic
interpretation of the human voice, based on models for phonemes which are the small
units of sound in any word and sentence. This technology has great potential for human
C EO
control of computers and machines, once the recognition rate and the speaker
independence can be further increased. Gestural control has been developed with music
technologies and in engineering. Currently, it is an emerging subject of research
including gesture sensing/capture technologies, intention/emotion detection, mapping
strategies, etc.
E –
H

The predominant form of control in human-machine interaction is the one using control
PL O

devices. Important types of control devices are keyboards (with alphanumeric or with
functional keys), mouse, joystick, trackball, steering wheel, pedals, knobs, and switches.
M SC

Force-displacement characteristics of these control devices need to be appropriately

designed. Also, some virtual control devices exist which can be operated on display
screens, e.g., buttons, sliders, and menu options.
SA NE

-
-
U

TO ACCESS ALL THE 31 PAGES OF THIS CHAPTER,

Click here

Bibliography

Alty J.L. and Johannsen G. (1989). Knowledge-based dialogue for dynamic systems. Automatica 25, 829
– 840. [This work deals with the use of artificial intelligence techniques and an interface paradigm based
upon the separation between dialogue and presentation].
Bainbridge, L. (1983). Ironies of automation. Automatica 19(6), Special Issue on Control Frontiers in
Knowledge Based and Man-Machine Systems (ed. A.P. Sage), 775 – 779. [This classical paper discusses

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

the ways in which automation of industrial processes may expand rather than eliminate problems with the
human operator].
Fejes, L., Johannsen G., and Strätz G. (1993). A graphical editor and process visualisation system for
man-machine interfaces of dynamic systems. The Visual Computer 10(1), 1 – 18. [This paper presents the
process visualization system of a human-machine interface and a graphical editor for designing dynamic
pictures of the process visualization system].
Helander M.G., Landauer T.K., and Prabhu P.V. (Eds.) (1997). Handbook of Human-Computer
Interaction (2nd Edition). Amsterdam: North-Holland. [This book is the most comprehensive
compendium that comprises all issues of human-computer interaction, also related to human-machine
interaction].
Johannsen, G. (1992). Towards a new quality of automation in complex man-machine systems.
Automatica 28, 355 – 373. [This work demonstrates the applicability of the human-centered approach
towards automation, by means of an integrated conceptual structure and a mathematical framework for
describing the human-automation synergism in technical systems].
Johannsen G. (1993). Mensch-Maschine-Systeme (Human-Machine Systems, in German). Berlin:
Springer. [This is a textbook that covers the whole breadth of human-machine systems, their foundations,

S
TE S
and many applications].

R
AP LS
Johannsen G. (1997a). Conceptual design of multi-human machine interfaces. Control Engineering
Practice 5(3), 349 – 361. [This paper deals with human-machine interfaces for cooperative supervision
and control by several human users, either in control rooms or in group meetings, and with the design
process for these interfaces emphasizing different design stages, user participation and knowledge-based
C EO
design support].
Johannsen G. (1997b). Cooperative human-machine interfaces for plant-wide control and communication.
Annual Reviews in Control 21 (ed. J.J. Gertler), 159 – 170. Oxford: Pergamon, Elsevier Science. [This
paper integrates achievements from the fields of human-machine and human-computer interfaces with
E –

considerations about more cooperation and communication].

H

Johannsen G. (2001). Auditory displays in human-machine interfaces of mobile robots for non-speech
PL O

communication with humans. Journal of Intelligent and Robotic Systems 32(2), 161 – 169. [This work
explains the design and the experimental investigation of auditory displays for mobile robots].
M SC

Johannsen G. (Ed.) (2002a). Analysis, Design and Evaluation of Human-Machine Systems 2001 (Proc.
8th IFAC/IFIP/IFORS/IEA Symposium, Kassel, 2001). Oxford: Pergamon, Elsevier Science. [This book
is one volume of a proceedings series of symposia on human-machine systems, sponsored by four major
international federations].
SA NE

Johannsen, G. (2002b). Human supervision and control in engineering and music – Foundations and
transdisciplinary views. Journal of New Music Research 31(3), Special Issue on Human Supervision and
Control in Engineering and Music (eds. G. Johannsen and G. De Poli), 179 – 190. [This paper provides an
overview on the field of the Special Issue and on all important aspects from the International Workshop
U

on “Human Supervision and Control in Engineering and Music” in Kassel, Sept. 2001. The main concept
of supervisory control is illustrated with engineering applications, as well as applied to music
performance].
Johannsen, G. (Guest Editor) (2004). Engineering and Music – Supervisory Control and Auditory
Communication. Special Issue, The Proceedings of the IEEE. 92(4), 583 – 758. [This Special Issue
includes selected papers mainly based on the International Workshop on “Human Supervision and
Control in Engineering and Music” in Kassel, Sept. 2001. Contributions from a transdisciplinary
approach of engineering and music are emphasized, particularly related to supervisory control and
auditory communication].
Johannsen G. (2008). Foundations of Human-Computer and Human-Machine Systems. Boca Raton, FL:
CRC Press - A Taylor & Francis Company. [This textbook is under preparation. It will be broad in scope
and deep enough for advanced undergraduates as well as graduates and practitioners. All fields, mainly
Human-Machine Systems Engineering, Human-Computer Interaction, and Human-Robot Communication
will be re-united and jointly presented as a coherent discipline].

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

Johannsen G., Ali S., and van Paassen R. (1997). Intelligent human-machine systems. Methods and
Applications of Intelligent Control (ed. S. Tzafestas), 329 – 356. Dordrecht: Kluwer. [This book chapter
presents an overview on knowledge-based support systems and knowledge visualization in human-
machine interfaces].
Johannsen G., Heuer J., and Tiemann M. (1997). Participative user interface design for industrial
processes. In: Proc. Third Silicon Valley Ergonomics Conference (ErgoCon 97), Palo Alto, CA, pp. 21 -
28. [This work illustrates participative design of human-machine interfaces with results from two research
and evaluation projects].
Rasmussen, J., Pejtersen, A.M., and Goodstein, L.P. (1994). Cognitive Systems Engineering. New York:
Wiley. [This book is a comprehensive guide to the field of cognitive systems engineering, with the
emphasis on analytical techniques for the design of all crucial aspects in human-work interaction].
Reason J.T. (1990). Human Error. Cambridge: Cambridge University Press. [This book presents the
background, the mechanisms, and the consequences of human error, with an appendix of six case studies
of accidents in complex high-technology systems].
Röse, K., Liu, L., and Zühlke, D. (2002). Human-machine-interaction design in Mainland China: Selected
cultural specifics. In: Analysis, Design and Evaluation of Human-Machine Systems 2001 (ed. G.

S
TE S
Johannsen), (Proc. 8th IFAC/IFIP/IFORS/IEA Symposium, Kassel, 2001), pp. 17 – 22. Oxford:

R
Pergamon, Elsevier Science. [This paper proposes an effective approach to elicit user requirements on

AP LS
culture-specific design issues and summarizes some cultural specifics for Mainland China].
Rouse, W.B. (1991). Design for Success: A Human-Centered Approach to Successful Products and
Systems. New York: Wiley. [This book provides a comprehensive, methodological framework for the
C EO
human-centered design of complex systems and covers the entire product life cycle].
Sheridan T.B. (1992). Telerobotics, Automation, and Human Supervisory Control. Cambridge, MA: The
MIT Press. [This book covers the theoretical and technological aspects of telerobotics, supervisory
control, and human-machine cooperation in the control of complex systems].
E –

Sheridan, T.B. (2002). Humans and Automation: System Design and Research Issues. New York: Wiley.
H

[This book is a tutorial for professionals and students who want to understand the many human interactive
PL O

aspects of automation].
Sundström G.A. and Salvador A.C. (1995). Integrating field work in system design: A methodology and
M SC

two case studies. IEEE Transactions on Systems, Man and Cybernetics 25, 385 – 399. [This paper
describes a framework for modeling information and knowledge types which end-users need].

Biographical Sketch
SA NE

Gunnar Johannsen is Professor of Systems Engineering and Human-Machine Systems – in the Institute
for Measurement and Automation of the University of Kassel, Germany, from 1982 to 2006. Since he
retired from his duties in teaching and administration in 2006, he concentrates more on research and,
U

particularly, on scientific writing. Further, he is orchestral conductor of Vienna classical and of

contemporary music since 1999.
He received his Dipl.-Ing. degree (1967) in communication and information engineering and the Dr.-Ing.
degree (1971) in flight guidance and manual control from the Technical University of Berlin, Germany.
In addition, he studied music for three years within the sound engineering curriculum at the University
School of Music, Berlin. In 1980, he habilitated (Dr. habil.) at the Technical University of Aachen,
Germany. From 1971 to 1982, he was division head in the Research Institute for Human Engineering near
Bonn, Germany.
During longer research stays, he worked in the University of Illinois at Urbana-Champaign in the USA
(1977–1978), in the Kyoto Institute of Technology (1995) and the Kyoto University (1995 and 2004) in
Japan, in the Technical University of Vienna, Austria (1999), also supported by the Institute of Electro-
Acoustics, Experimental and Applied Music of the University of Music and Performing Arts Vienna, as
well as in the University of British Columbia in Vancouver, Canada (2004).
Gunnar Johannsen is author and editor of numerous publications in conference proceedings, refereed
journals, and in book form, e.g., author of “Mensch-Maschine-Systeme” (Human-Machine Systems;

©Encyclopedia of Life Support Systems (EOLSS)

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION – Vol.XXI - Human-Machine Interaction - Gunnar Johannsen

Springer, Berlin 1993). In IFAC (International Federation of Automatic Control), he founded and chaired
the Working Group on Man-Machine Systems (1981–1990). He served as Chairman for four of its
Symposia between 1982 and 2001. Also in 2001, he created and was main responsible of an International
Workshop on Human Supervision and Control in Engineering and Music, with an embedded orchestra
concert in which he conducted Takemitsu’s November Steps.
He is Fellow of the IEEE (Institute of Electrical and Electronics Engineers; Systems, Man, and
Cybernetics and Computer Societies) “for contributions to human-machine systems engineering,
cognitive ergonomics, human-computer interface design, and human-centered automation”. Further, he
was recipient of a Japanese-German Research Award (granted by JSPS, the Japan Society for the
Promotion of Science). In 2005, he received the title of Docteur Honoris Causa (Dr. h.c.) from the
Université de Valenciennes et du Hainaut-Cambrésis in North France.
His current research interests are in human-machine systems, human-centered computing, automation,
and design, cognitive systems engineering, graphical, auditory, and multimedia user interfaces, gestural
control, audio and music technologies, decision support systems, knowledge engineering, usability
engineering, co-operative work, and interrelations between engineering and music.

S
TE S
R
AP LS
C EO
E –
H
PL O
M SC
SA NE
U

©Encyclopedia of Life Support Systems (EOLSS)