Методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами
The problems of intelligent control system organization are considered: determining the number of intellectualization levels, the sequence of actions required for analysis of the control process, adding to the control system new elements providing for enhancement degree of its intellectualization, s...
Збережено в:
| Дата: | 2022 |
|---|---|
| Автори: | , , , |
| Формат: | Стаття |
| Мова: | Англійська |
| Опубліковано: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2022
|
| Теми: | |
| Онлайн доступ: | https://journal.iasa.kpi.ua/article/view/275020 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | System research and information technologies |
| Завантажити файл: |  |
Репозитарії
System research and information technologies| _version_ | 1866302889881763840 |
|---|---|
| author | Melnykov, Serhii Malezhik, Petro Gasanov, Aydin Bidyuk, Petro |
| author_facet | Melnykov, Serhii Malezhik, Petro Gasanov, Aydin Bidyuk, Petro |
| author_sort | Melnykov, Serhii |
| baseUrl_str | http://journal.iasa.kpi.ua/oai |
| collection | OJS |
| datestamp_date | 2023-05-21T20:04:38Z |
| description | The problems of intelligent control system organization are considered: determining the number of intellectualization levels, the sequence of actions required for analysis of the control process, adding to the control system new elements providing for enhancement degree of its intellectualization, special features of its structural organization, estimating the possibilities of intellectualization, providing examples of practical intellectualization. The primary purpose of the study is to determine the purposeful organization of intelligent control systems as well as the necessity and usefulness of systemic consideration that takes into consideration the following: requirements of the problem statement, characteristics of the environment, means for acquiring and processing necessary information, working control mechanisms, functional characteristics and experience of user-operator. As a result of the analysis performed, characteristic levels of the intellectual development of a system were determined, the stages of performing intellectualization of a control system were proposed, and the effectiveness of proposed solutions for practical problems was shown. |
| doi_str_mv | 10.20535/SRIT.2308-8893.2022.4.04 |
| first_indexed | 2025-07-17T10:28:05Z |
| format | Article |
| fulltext |
S.V. Melnykov, P.M. Malezhyk, A.S. Gasanov, P.I. Bidyuk, 2022
44 ISSN 1681–6048 System Research & Information Technologies, 2022, № 4
UDC 629.7.05
DOI: 10.20535/SRIT.2308-8893.2022.4.04
METHODOLOGICAL ASPECTS OF OPERATIVE CONTROL
SYSTEM INTELLECTUALIZATION FOR DYNAMIC OBJECTS
S.V. MELNYKOV, P.M. MALEZHYK, A.S. GASANOV, P.I. BIDYUK
Abstract. The problems of intelligent control system organization are considered:
determining the number of intellectualization levels, the sequence of actions re-
quired for analysis of the control process, adding to the control system new elements
providing for enhancement degree of its intellectualization, special features of its
structural organization, estimating the possibilities of intellectualization, providing
examples of practical intellectualization. The primary purpose of the study is to de-
termine the purposeful organization of intelligent control systems as well as the ne-
cessity and usefulness of systemic consideration that takes into consideration the fol-
lowing: requirements of the problem statement, characteristics of the environment,
means for acquiring and processing necessary information, working control mecha-
nisms, functional characteristics and experience of user-operator. As a result of the
analysis performed, characteristic levels of the intellectual development of a system
were determined, the stages of performing intellectualization of a control system
were proposed, and the effectiveness of proposed solutions for practical problems
was shown.
Keywords: control system, intellectualization, organization of control systems, in-
telligent control, man-machine systems.
INTRODUCTION
The urgency of the problem of control system intellectualization is determined by
the modern requirements to development and implementation new control sys-
tems for dynamic processes and systems that could use large volumes of informa-
tional data and are capable to perform intellectual processing of the data to reach
high quality control in real time. Dynamic modern development of the computa-
tional means and particularly the means for parallel data processing like graphical
and tensor processors as well as new intellectual data processing methods create
favorable conditions for reaching high quality control. Together with the possi-
bilities mentioned the requirements regarding data optimization are preserved
aiming to determining the confidence intervals and characteristic numbers.
The complex approach to taking into consideration the purposeful functioning
of a system, to determining dimensionality of complex systems characteristics and
control environment, to description of functioning machine operator provides
creating conditions in the direction mentioned.
The trend to automatizing of all sides of human activities (transportation, in-
dustrial production, agriculture, economy, home activities, social sphere etc.) re-
quire development of high-tech systems with intellectual functions, capable to
interact with distributed data systems and use remote services.
The paper contains methodological aspects and examples of organizing intel-
lectual control systems for complex dynamic objects and processes. It is also pro-
posed the methodology for organizing control systems with functional structures
that implement some separate characteristics of intellect.
Methodological aspects of operative control system intellectualization for dynamic objects
Системні дослідження та інформаційні технології, 2022, № 4 45
Intellectual technology, intellectual control, intellectualization – all these no-
tions are mostly accepted intuitively. When transition occurs from some purpose-
ful functioning of a system to the process that exhibits intellectual features, then
different subjective opinions come to being that gradually transform into the pro-
cesses of search for high quality solutions and more complicated tasks.
The considerations proposed in the study are linked to emergence and
development of natural evolutional intellect but the possibilities are preserved for
hybrid combinations with other approaches. The intellectual level of a control
system is linked to effectiveness of its functioning but there is no one-to-one
correspondence. The general estimate of effectiveness of the system created and
the measure of its intellectualization is rather relative and can be determined via
investigating of its functioning using various methods in natural conditions or via
computer simulations. The qualitative estimation of control system intellectuality
supposes availability of some definite structure as well as adaptation of the struc-
ture and model parameters, memorizing control actions and their results, the pos-
sibilities for constructing models of environment and the system under study etc.
The quantitative estimation of intellectuality can be expressed by the number of
micro-situations in which within limited time can be found solution for stated
problem. Time is a critical factor that influences general effectiveness of practical
functioning of a system being studied. The effectiveness of a system is often con-
sidered as necessary initial but not sufficient condition of intellectuality. When
strict conditions on the decision making time exist the effectiveness of general
functioning of a control system that has lower intellectuality level may become
higher. Certainly here we have to higher numerical effectiveness criterion.
PROBLEM STATEMENT
The use of conforming to the laws of evolutionary development biological sys-
tems for constructing engineering systems is carried out during all history of
mankind. Development and improvement of intellectual functions of biological
system stands in the same position, i.e. this is not exception. Here it is proposed to
use basic natural conforming to the laws intellectual developments for creating
technical systems though we don’t remove the possibilities of using effective en-
gineering solutions in hybrid systems. Development of such engineering systems
requires paying special attention to selecting and processing only necessary in-
formation for solving specific goal problems and further usage of the data re-
ceived.
One of key moments of carrying out analysis while solving complex practi-
cal problems and touching upon determining characteristics related to confidence
interval and control of dynamic objects is taking into consideration and coordina-
tion of used environment characteristics and the controlled object with the goal
requirements and possibilities of receiving information from gauges that is limited
in its precision in noisy conditions as well as taking into consideration discrete
time intervals for making control decisions. As an example, the problem is con-
sidered to control flying object in such a way to lead it into given limited space
from the space of initial states. Taking into consideration of such requirements
provides the possibility for carrying out micro-situational analysis and find opera-
tive control, to determine the areas of reachability for the goal values of variables
and parameters for various control algorithms, and make conclusions regarding
effectiveness of the usage for the algorithms applied.
S.V. Melnykov, P.M. Malezhyk, A.S. Gasanov, P.I. Bidyuk
ISSN 1681–6048 System Research & Information Technologies, 2022, № 4 46
Today there exist a large number of studies directed to description of intel-
lectual features of biological systems and their implementation in technical sys-
tems [1, 2]. Of particular popularity and success today has the use of analogs for
neural nets [3]. Practical implementation of intellectual systems is performed in
the most diverse spheres of human activities; it is especially topical and substanti-
ated problem for the unmanned transportation systems [4]. The problem of practi-
cal implementation of the intellectual systems is mainly related to the high dimen-
sion of input data, availability of noisy components, uncertainties in internal and
external parameters of control processes as well as in absence of effective basic
control solutions and structures that could be modernized. The systematic consid-
erations of forming intellectual features of biological organisms helps to over-
come the difficulties mentioned in the practice of creating and implementing in-
tellectual technical systems.
Using quite generalized considerations it is possible to analyze the process of
forming some system with control structures and control object in chosen goal
environment as gradual refinement of the system organization.
The system that performs definite goal task is functioning in a specific se-
lected environment and basically presents a part of the environment, and it can
serve as external sphere for some other systems. The structural division of the
sphere and system, of an object and control system can be quite conditional, that
could be used for logical understandable description of the processes related to
functioning of the system.
Selection of the goal environment is performed with taking into considera-
tion the specificity of the problem being solved. Out of the world where every-
thing is happening the part of environment is selected in which the specific task is
implemented. Then the dimensionality of substantial characteristics and states of
the goal environment is determined, selection of the controlled object, control sys-
tem, input and output data, conforming to the laws functioning of a man in the
process of search the goal solutions. Then the goal situations are described and
determining of the sets of goal micro-situations that create situations, and it is de-
termined the goal functioning in the situations and micro-situations.
The search of decisions necessary for reaching the goals determined in se-
lected situations and micro-situations can be characterized as determining the
function of data for output variables at the inputs of the goal (under investigation)
system (for example, control system or decision support system). Taking into
consideration the evolutional development of biological systems determine the
basic directions of their improvement and further development in the sense of en-
hancement of the level of intellectual possibilities without strict division on
availability or absence of intellect. In the problems that are solved by simple
biological organisms the similarity of intellectual features can be highlighted by
the specificity of subsystems receiving and processing external information
(intellectual sensors). Further enhancement of the intellectuality level is linked to
constructing of environment models and models of possible behavior of an object
(being studied) in this environment with estimation of final results.
When intellectual control system is synthesized the attention should be paid
to the urgent methodological questions related to selection and processing input
information. Here also the aspects of constructing distributed control structures
and databases as well as the questions of decision forecasting in the nonlinear
non-stationary environments should be taken into consideration.
Methodological aspects of operative control system intellectualization for dynamic objects
Системні дослідження та інформаційні технології, 2022, № 4 47
THE CONCEPTUAL IDEAS REGARDING ORGANIZATION OF
INTELLECTUAL CONTROL SYSTEMS
When intellectual control system is synthesized it is necessary to take into consid-
eration a series of general structural and functional norms. The norm touching
upon adequate reflection of features of living beings and a man in the structure
and functions of modules and systems being developed should be necessarily con-
sidered. The basic structural and functional similarity is reflected through devel-
opment of logic-and-dynamic models of control processes. Extended functional
similarity is reflected through constructing the models of adaptive and optimiza-
tion stochastic transforms of control processes.
Consider the idea of compact discrete representation of the recognition
processes and object control. Such representation supposes selection of
substantial variables, forming of input and output spaces of features, the measure
of discretization, and optimal division for specific control environment.
Now consider the idea of finding compact transforms that provide for opti-
mal solution and correspond to representations of man-operator and experts. In
particular, here it is necessary to perform processing of several data streams with
their mutual comparison in functional space, and perform the goal transforms
with several levels of decision making based only on the short time forecasts, but
also estimate development of situation on the long-term time intervals.
The intellectual integrated control represents multicomponent process of
searching for solutions that supposes forming of a series of goal images for con-
trol object and control environment, their further refinement, transforming and
consolidation in the generalized goal space.
The control process is related to discrete or situational ones. The intervals of
discretization or situations are determined by the interactions of control object and
control subject as well as influence of environment. Here the most substantial in-
fluence on the discretization of control process makes man-operator and corre-
spondingly his characteristics and practical experience. The next substantial rea-
son for control process discretization is availability of noisy components in the
control system and environment. Another factor influencing discrete characteris-
tics of the control process is related to bad quality of data characterizing initial
and current characteristics of controlled object and control environment. A spe-
cific feature of this representation is also taking into account of interaction
between man-operator and controlled object even if the control is performed in
fully automatic mode. Here automatic control is constructed with taking into
consideration of man-operator representation about possible control process and
his experience of work in conditions of preserving monitoring functions and
correction of possible control decisions. Thus, the decisions that are performed
(implemented) are directly or indirectly related to representation of man-operator
and his conformity of laws regarding information transforming. Here it is of
particular importance for development of separate optimal systems have the goals
set by man-operator and their actual practical implementation.
The process of system intellectualization is considered as gradual enhance-
ment of its possibilities regarding effective goal stated functioning on the ex-
tended set of environment states. We can state that when the control system
reaches some goal using its actions more effectively than without such activity,
then there exists some level of its intellectuality.
As an example of possible representation of the development levels for intel-
lectual characteristics of automatic and automatized control systems the following
levels are given:
S.V. Melnykov, P.M. Malezhyk, A.S. Gasanov, P.I. Bidyuk
ISSN 1681–6048 System Research & Information Technologies, 2022, № 4 48
1. The intellectualization level at which preliminary data processing is per-
formed in a single control channel. For example, it can be the case of using spe-
cialized goal sensor or the system based upon directive control.
2. The intellectualization level at which several information channels are
working and extra preliminary data processing is performed together with selec-
tion of substantial data necessary for reaching the goal. It is also performed identi-
fication and taking into account of possible uncertainties.
3. The intellectualization level at which development of information trans-
formers is performed that take into account stable conformity to the laws of man-
operator activities like logical inference or stereotype behavior.
4. The level of intellectualization at which adaptation is performed to the
specificity of accepting and search for solution of the specific operator, the
generation takes place and expansion of possibilities for selecting alternative in-
formation transforms.
5. The level of intellectualization at which during the control processes is
taken into account the set of data- and knowledge bases about control object and
environment; the adequacy is observed about system functioning according to
basic notions and criteria with the man-operator what can be considered as initial
level of image-like control.
6. The level of intellectualization at which exists adequacy of control system
functioning according to the image representation of situations by the man-
operator and decision forming by the technical part of the system in the definite
specific control area.
7. The intellectualization level at which there exists adequacy of system
functioning according to the basic information data (and outside of separate spe-
cific control area). Here is performed in concordance with multiple situations in-
teraction of a man-operator with the engineering part of the control system.
8. The level of intellectualization at which the systems under study exhibit
the characteristics of creative development and refinement.
ORGANIZATION OF GOAL REACHING SYSTEMS
The conformity to the laws for organizing goal reaching systems is expressed in
structural and functional characteristics of highly organized living beings [5].
Analysis of these characteristics allows for determining the following basic con-
formities to the law.
Availability of some time and space continuum for the control system exis-
tence in the form of a part of selected environment.
Availability of definite levels of external and internal activities, and freedom
of the system under study.
Availability of signals of definite levels for control and interaction of the
system with environment and control of internal states.
The number of distinctive signal levels allows for selection of a limited
number of possible states in the system and states of the system in the environment.
Availability of initial structural organization that determines initial choice of
possible system reactions to external and internal influences.
Availability in the initial structure of motivation basis (for example, for
satisfying of the life necessities as well as role necessities and self-development)
that is expressed in the form of definite behavior or program of actions.
Availability in the initial structure of plastic elements that allow for re-
cording stimuli signals, emergence of initial reactions and results of system inter-
action with the environment.
Methodological aspects of operative control system intellectualization for dynamic objects
Системні дослідження та інформаційні технології, 2022, № 4 49
Availability of hierarchical structural system organization that determines in-
teractions of its basic structures, motivation basis, plastic elements and the ele-
ments that are responsible for actualization of system reaction.
An example of the sequence of organization and search for the solutions in
goal set man-machine systems:
General formulation of the goal settings.
Selection of the goal space characteristics and control object.
Activities analysis of a man-operator under realization of various general
and specific goal settings.
Analysis of selected environment in the space of goal set functioning of a
system – goal set reduction of the control environment.
Analysis of the control object in the space of goal-set functioning of a
system – the goal set dimension reduction of control object.
Creation of a basic structure of a control system in realization of goal set-
tings.
Distribution of information and control system resources directed to ef-
fective and reliable realization of goal settings.
Creating of possibilities for database and knowledge base expansion, ad-
aptation, learning and repeated learning while realizing of the goal settings.
Multilevel integration of resources and structures for generating decisions.
Reaching integrated functioning of technical part of a control system and
man-operator.
Introducing into integrated goal system the features of creative developments.
ORGANIZING DISTRIBUTED REMOTE CONTROL SYSTEMS
Development of effective distributed remote control systems includes the fol-
lowing stages [6].
The problem statement formulation for remote control and determining re-
quirements to the functional characteristics of control object and control system itself.
Formulation of specific temporal requirements to functional agents that are
selected in the functional structure of control system under study.
Determining of characteristics for the existing structure of the net that can be
used for solving the specific problem of related to remote control.
Determining of time delays emerging during the use of different protocols
for information exchange within the net hired.
On the basis of the analysis performed distribution of the control blocks
structure at the remote object is performed and in the control system.
The functional optimization is performed of control for the remote system at
the expense of extra knowledge about characteristics of the system and environ-
ment.
Adding to the net structures of new information devices for receiving and
processing information.
Creating of a developing net infrastructure with adding new intellectual fea-
tures and possibilities.
Extra development of net information exchanging protocols and organization
of new channels for information input.
For effective functioning of control systems with distributed information and
executive resources in multitask net structures it is necessary to perform complex
S.V. Melnykov, P.M. Malezhyk, A.S. Gasanov, P.I. Bidyuk
ISSN 1681–6048 System Research & Information Technologies, 2022, № 4 50
integrated considering of all elements of the control process and taking into ac-
count of the following recommendations:
– distribution of control information according to its value regarding the con-
trol goals and acceptable transmission delays;
– determining of specifications for data transmission protocols and for dif-
ferent types of information;
– determining expediency of usage for different channels and routs of infor-
mation transmission;
– setting priorities of data meaningfulness and the sequences of its transmis-
sion in the information structures;
– creating the structures that provide for duplication or excessiveness of the
data being transmitted;
– determining of structure for distribution of functional modules of the sys-
tem and control object depending on possible intervals of control signal delay.
Besides taking into account of systemic requirements to information trans-
mission additional possibilities can be hired for organizing of control structures
that take into account perspective methods of control for high speed dynamic pro-
cesses and structures for distribution of control means on the object and control
system. All this helps to improve substantially quality of control for dynamic sys-
tems [7]. Fig. 1 shows examples of control organization for remote dynamic objects.
The complex usage of the methods developed for control of high-speed dy-
namic processes and taking into account of proposed recommendations regarding
organization of information transmission in the distributed net structures allows
for reaching general improvement of qualitative and quantitative effectiveness
measures in the process of remote system control.
THE SYNTHESIS OF GOAL REACHING CONTROL OF A FLYING OBJECT IN
CONFLICT AND ACCIDENT CONDITIONS
Consider example of constructing an airplane model and description of the air-
plane control in algorithms in the non-scheduled situations and equipment faults.
Of high importance for successful solving of a flight safety problem are the ques-
Control organization with
taking into account possible
delays in distributed
control systems
Shifting a part of
Man-operator dynamic
characteristics
to controlled object
Shifting a part
of intellectual
Man-operator
functions to
controlled object
Development
of the world model
on the controlled
object
Development of the
control team model
on the controlled
object
Shifting a part
of control system
functions to executive
mechanisms
Development
of remote object
model in distributed
control system
Development of
environment model
in distributed
control system
Development of the
Man-operator model
in distributed
control system
Fig. 1. Examples of control organization for remote dynamic objects
Methodological aspects of operative control system intellectualization for dynamic objects
Системні дослідження та інформаційні технології, 2022, № 4 51
tions of reliable and stable with regard to interferences of flight control when
various goal maneuvers are performed. Also important are the questions of devel-
opment control algorithms for providing the flights on an airplane with low stability
reserve or in the case of unstable airplane.
The control algorithms being developed have to enhance the plane “vitality”,
reliability of its control system when maneuvers are performed. These algorithms
also should provide for continuation of performing set tasks by an airplane when
one or several control mechanisms exhibit fault regime. This problem can be
solved at the expense of fault compensation by the control system thanks to re-
configuring of the control process by varying aerodynamic surfaces of an air-
plane.
Solving of the problems mentioned is proposed in the frames of a concept of
constructing distributed control system. It is supposed that an airplane has various
controlled aerodynamic surfaces, and control by executive mechanisms and sur-
faces can be performed separately. Flexible structure of control system provides
the possibility for changing its configuration during flight, and use various func-
tional regimes of aerodynamic surfaces to create aerodynamic forces of necessary
configuration.
To reach stable movement of an airplane, automatic controllers for tangent
angle, turn angle, and rove angle are used. The controllers should provide for
reaching by airplane of the set orientation angle, performing maneuvers and the
flight with given tangent, turn and roam angles. The automatic control of orienta-
tion angles provides for compensation of control mechanisms faults.
The algorithms development suggests on-board computer application in the
control system. The on-board computer coordinates the functioning of control
mechanisms and provides for performing in real time necessary computations and
generation of control commands.
To solve the problems of performing goal tasks computer simulation of the
control algorithms developed was performed for reaching the airplane control. To
solve the simulation problem the flight of aerodynamically unstable airplane had
been modeled in perturbed atmosphere. Simultaneously, it was modeled functioning
of control system for aerodynamic surfaces of the airplane. Using the model
created the testing of functioning of the algorithms developed was performed. The
model developed allows for performing simulation of various modes and stages of
flight as well as receive estimates of effectiveness of the algorithms functioning.
The simulation allows for receiving information about possibilities and areas of
application for each algorithm and control system as a whole.
Generally, the perturbed or controlled movement of an airplane is taking
place in three-dimensional space because here simultaneously are taking place
changings of parameters and variables that define direct and side movement.
When the amplitudes of parameter changings related to the movement of an
airplane are small then equations of direct and side movement of the plane can be
approximately studied separately. However, when the movement parameter
changings are quite large such separation of the equations into direct and side
movements starts to produce unacceptably large errors, and it is necessary to ana-
lyze complete system of equations describing spatial movement.
Development of the airplane dynamics model was performed on the purpose
of simulation of various flight regimes and functioning of control system. To
reach this goal complete system of equations describing spatial movement of an
airplane was used.
S.V. Melnykov, P.M. Malezhyk, A.S. Gasanov, P.I. Bidyuk
ISSN 1681–6048 System Research & Information Technologies, 2022, № 4 52
The airplane movement takes place under influence of aerodynamic forces
and moments produced by engines and gravitation forces. Generally, to define an
airplane movement it is necessary to solve the following problems [8]:
– to find angular and linear velocities of an airplane movement, induced by
the influence on it of all mentioned forces and moments;
– to determine the angles of the plane orientation relatively the flow that
meet it, and coordinate axis linked to the Earth;
– to determine the shift of an airplane with respect to Earth.
Each of these tasks is supported by special set of differential equations. The
model uses quite complete and precise equations of an airplane movement for
performing necessary computations and allows for achievement of truthful results
for modeling control system functions. In this case an airplane is considered as
absolutely rigid body
The mass and inertia moments of airplane are considered as unchangeable
during the time of modeling the plane and correspond to initial state of equilib-
rium flight.
It is also suggested that the plane configuration has symmetry plane and its
masses are distributed symmetrically with respect to this plane.
To represent the movement model of unstable flying apparatus in three-
dimensional space the work regarding development of stabilizing algorithms of
the apparatus has been carried out with taking into consideration cross-channel
links in the maneuvers are performed it is supposed that various external
disturbances exist such as wind, atmosphere turbulence etc., as well as internal
ones, such as poor precision of computations and sensors. To construct the
trajectories of given movement into the goal area it is necessary to take into
consideration the most substantial information characteristics of control object
and environment, and have the possibility of continuous correction of the trajec-
tory due to influence of random factors. Taking into account the general
considerations given above, now present an example of control organization for
flying apparatus at the stage of landing in accidental conditions and with
influence of random wind changings.
Airplane control during the landing stage is the most complicated task of its
control. That is why the problems of automation of airplane landing are given
substantial attention. The system providing for the airplane landing for the case of
engine fault has been developed as an compound part of integrated control system
for complex dynamic objects that contains a program simulating movement of a
flying apparatus in three-dimensional space.
The airplane landing in accident mode of engine fault puts ahead extra re-
quirements to organization of functioning of interconnected system including pi-
lot – automated control system – flying apparatus. Such systems, oriented to ac-
tive interaction with man-operator in the extreme situation, should take into
account specific features of the man-operator activities in such situations as well
as accepting of incoming information by the man. This problem can be formu-
lated as mutual coordination of goals and structure between automated part of the
control system and activities of the man-operator. Such considering of a control
system as integrated man-machine system can serve as a ground for determining
general structure of control and for optimization of respective structural and func-
tional solutions.
In the process of the system development it was accepted that man-machine
system for controlling the object in critical regime of functioning requires taking
into consideration for effective and reliable reaching the goals of control the fol-
lowing factors [9]:
Methodological aspects of operative control system intellectualization for dynamic objects
Системні дослідження та інформаційні технології, 2022, № 4 53
– activity characteristics of a man-operator, special features of the process of
accepting and processing the information by the man-operator as well as generat-
ing control actions;
– flexible structure of the algorithms developed and coordination of the algo-
rithms structures with stable stereotypes of the man-operator;
– the possibility of operative switching the control functions between man
and automaton;
– taking into consideration and compensation of various types and origin dis-
turbances, dynamic analysis of situations, correction and synthesis of trajectories.
Computer simulation of the control algorithms developed using realistic air-
plane and environment model provided for the possibility of estimating effective-
ness of reaching final goal in complicated navigation environment. For the task of
emergency landing of a cargo plane it is possible to generate substantial en-
hancement of dimensions for the initial states area from which positive result can
be reached in comparison with known trajectory control algorithms.
It was also developed and optimized regarding the types and content of con-
trol information the interface providing for interaction of a pilot with control sys-
tem. The interface was coordinated with characteristics of the man-operator and
his information requirements in critical situations.
ESTIMATION OF CONTROL EFFECTIVENESS
Estimation of decision effectiveness in general case should be received for the
whole region of allowable initial conditions. As a rule the region is rather large
and simple analysis (trying) of all points or variants requires substantial time and
appropriate computational means. Taking into account the systemic methodology
for specific realizations of a control process, generally, we managed to reduce the
number of variants for determining effectiveness of analyzed decisions at the ex-
pense of considering continuity of the control process and conformity to the laws
of systemic interactions. Continuity, as well as stability of the nature of control
processes suggests that after changings in external conditions that influence the
changings of object states, it is occurring relatively small change of final control.
Usually, these values (changings) do not exceed some definite threshold in the
limits of one given macro-situation. On the limit transitions between the situations
(transition micro-situations) the
control may change in the uneven
form, and the changings in final
results can substantially distinguish
from inter-situational values. As an
example of calculating such esti-
mates it was considered control of
the goal movement of an airplane
with constraints regarding energy
possibilities of the apparatus in per-
turbed environment. On the basis of
control methodology considered
above it was created a series of ef-
fective control systems for aviation
and marine transportation means in
complex navigation conditions, in
conditions of equipment fault and
strong influence of external distur-
bances.
W
X0
XC
YC
Fig. 2. The region of initial states for flying
apparatus, from which successful landing is
performed. The height echelon — 5000 m.
Wind is favorable: X0 — initial state of an appa-
ratus; W — state of wind on the landing strip;
XC, YC — coordinate directions
S.V. Melnykov, P.M. Malezhyk, A.S. Gasanov, P.I. Bidyuk
ISSN 1681–6048 System Research & Information Technologies, 2022, № 4 54
Figs. 2–4 show examples of reachability regions for effectiveness estimating
of control algorithms for cargo airplane, where the arrow indicates to initial
course of flying apparatus, and sharp angle of the triangle points to the wind di-
rection. The computer simulation of the control algorithms developed with realis-
tic model of the airplane allowed for estimate effectiveness of reaching final goal
in a complex navigation environment. The area of the initial positions of the
airplane, from where a positive landing result is achieved, can be increased up to
2.5 times in comparison with the known methods of landing along a
predetermined trajectory.
DIRECTIONS FOR PERSPECTIVE RESEARCH
The modern technical level of information processing tools and the experience of
creating control systems capable of operating in automatic mode and effectively
interacting with a person in critical situations makes it possible to identify
promising areas of research and development.
It is proposed development of systems that include the sense notions, data-
bases and knowledge bases, objects (systems) and their interactions (selected or
general model of the world).
It is also proposed development of basic elements and processors that pos-
sess the features of adaptation, learning and repeated learning for organizing par-
allel data processing in stochastic environment.
We propose development of interaction architecture revealing actual interac-
tion for modern computational means, basic elements and stochastic environment.
Development of structure and separate components for integrated and dis-
tributed control systems.
Reaching necessary level of automation various purpose of distributed data-
bases and knowledge bases.
It is necessary to reach high quality of automatic and automated remote con-
trol for dynamic objects in real time.
W XC
YC
X0
Fig. 3. The region of initial states for
flying apparatus, from which successful
landing is performed. Height echelon —
5000 м. Wind unfavorable: (X0 — ini-
tial state of an apparatus, W — state of
wind on the landing strip, XC, YC —
coordinate directions)
Fig. 4. The region of initial states for flying
apparatus, from which successful landing is
performed. Height echelon — 3000 m. Wind
unfavorable: X0 — initial state of an appara-
tus; W — state of wind on the landing strip;
XC, YC — coordinate directions
W XC
X0
YC
Methodological aspects of operative control system intellectualization for dynamic objects
Системні дослідження та інформаційні технології, 2022, № 4 55
Enhancement of intellectual possibilities of control systems and decision
support systems with making use of distributed human and machine resources.
It is proposed development and application of hybrid man-machine intellec-
tual complexes and systems for processing databases and knowledge data.
These are quite general directions that acquire in sophisticated up-to-date
conditions special meaning. Development of multiprocessor computational means
and cloud computing, storing knowledge and accessibility to large databases re-
lated to control systems and environments with making use of remote access pro-
vide for the new possibilities of creating modern control system with features of
artificial intelligence. It is particular importance development and practical use of
neuron-like structures and technologies of convolutional type and deep learning
[3, 10, 11]. The active use of accessible databases related to various object types
for neuron-like systems learning provided the possibility for getting results starting
from 2014 of automatic object recognition that overcome the results of recogni-
tion received by man. On the basis of similar technologies new autonomous con-
trol systems are developed for transportation means, for machine translation using
different languages, the systems for searching objects in the images and genera-
tion of their descriptions, the systems for autonomous carrying out dialogs in
separate areas of activities, the systems for synthesis of stylized images, and many
of other new developments.
CONCLUSIONS
In the process of organization of effectively functioning man-machine control
system it can be distinguished the following basic stages of development control
algorithms and structures.
First is the stage of creating adequate model of object and control environ-
ment. Here, it is necessary, as possible, to develop precise enough models. The
limiting factors usually are the characteristics of environment noise, low observa-
tion precision of object variables, and impossibility to reach high precision of rep-
resentation for complicated functional characteristics, available, for example, in
the form of approximate table values.
On this stage the limit possibilities of a control object as a whole are studied,
and the problems are solved touching upon stabilization of separate characteristics
and development of automatic subsystems for them.
The second stage is usually devoted to development of basic control struc-
ture that is characterized by high stability and reliability of functioning. The struc-
ture has its specific logic of functioning for the basic set of realizable situations,
logic of carrying out testing of basic nodes and logic for the usage of reserved and
autonomous adaptive equipment.
The third stage includes development of means for representing control in-
formation and the means for monitoring functioning of the control system as a
whole. For the second and third stages it is particularly important taking into con-
sideration the goal requirements of a man-operator, his knowledge of functional
characteristics of the system and the use of his former experience of work with
typical control systems of definite class.
The fourth stage is devoted to implementation of extended matching of
operator and control system in conditions of higher risk and non-trivial situations.
S.V. Melnykov, P.M. Malezhyk, A.S. Gasanov, P.I. Bidyuk
ISSN 1681–6048 System Research & Information Technologies, 2022, № 4 56
Besides, the possibilities are created for extended situational analysis and mutual
adaptation between man-operator and control system being created.
The fifth stage is directed to providing for the control system with the means
of interaction with man-operator using the language of natural representations and
notions, providing the system with possibilities of realization creative elements of
the control process.
The indicated general ideas regarding organization of intellectual control
systems receive extended extra possibilities of implementation in modern engi-
neering and informational conditions. Correct discretization of the information
receiving processes, of the information transforming and control processes should
be matched to the goal tasks of the system and interactions with man-operator. A
detailed analysis of control processes on the relatively short time intervals (micro-
situations) is coordinated well with the development of modern neural like sys-
tems in the form of multilayer high precision networks.
The appearance of new computational possibilities in the form of tensor pro-
cessor and cloud services (computing), that provide for an access to such equip-
ment, create new possibilities for introduction of intellectualization to information
transformations and, consequently, constructing of intellectual control systems.
The practical development results are concentrated on control of moving dy-
namic systems, more precisely, on flying objects in critical regimes of flight.
It was also considered the problem of directing flying apparatus into given
limited region with predetermined state characteristics in conditions of availability
substantial external disturbances and limited control resources. The three-
dimensional computer simulation model, environment and control algorithms
were proposed that are distinguished by the high similarity measure with actual
functioning of practically available systems.
The control algorithms developed are related well to the experience of pilots
regarding control of flying apparatus in conditions of strong disturbances of
various nature and with availability of substantial uncertainties in information
data. These algorithms allow for making control decisions at each moment,
construct optimal goal trajectory with taking into consideration current state
characteristics of controlled object and environment.
The use of the micro-situational analysis and synthesis methods allows for
automation in different spheres of practical human activities, especially in the di-
rections where exist substantial uncertainties of data, and deficit of time for deci-
sion making.
The results achieved are directed to development of information technolo-
gies for the goal reaching systems of various classes, and have high practical
meaning for solving wide class of problems regarding control of complex dy-
namic systems in the conflict and accident conditions. The results can also be ap-
plied in the analysis and anticipating of catastrophic events in economic and so-
cial systems and processes, in organizing of multilevel industrial productions in
conditions of uncertainty and deficit of resources.
REFERENCES
1. M. Zgurovsky and Yu. Zaychenko, Fundamentals of computational intelligence:
System approach. Springer, 2016, 275 p.
2. Max Tegmark, Life 3.0: Being Human in the Age of Artificial Intelligence. Penguin,
2018, 380 p.
Methodological aspects of operative control system intellectualization for dynamic objects
Системні дослідження та інформаційні технології, 2022, № 4 57
3. Ian Goodfellow, Yoshua Bengio, and Aaron Courville, Deep learning. Cambridge,
MA, USA: The MIT Press, 2016, 800 p.
4. R.W. Beard and T.W. McLain, Small Unmanned Aircraft: Theory and Practice.
Princeton, NJ: Princeton University Press, 2012, 320 p.
5. S.V. Melnykov, “Conformity to the laws of organizing control in ergatic systems,” in
Cybernetics and computers, issue 155, pp. 3–15, 2007.
6. S.V. Melnykov, “Conformity to the laws of organizing complex ergatic systems and
creation of structure for remote control of dynamic objects,” in Cybernetics and
computers, issue 168, pp. 70–79, 2012.
7. S.V. Pavlova, Yu.P. Bogachuk, S.V. Melnykov and A.Yu. Gospodarchuk, “Organiz-
ing of effective data exchange in the network control systems for dynamic objects,”
in Cybernetics and computers, issue 177, pp. 43–53, 2014.
8. G.S. Bushgens and R.V. Studnev, Dynamics of airplane. Dynamics of direct and side
movement. M.: Mashinostrojenie, 1979, 372 p.
9. V.V. Pavlov, Synthesis of strategies in man-machine systems. Kyiv: Vyshcha shkola,
1989, 162 p.
10. Neil Wilkins, Artificial Intelligence: An Essential Beginner’s Guide to AI, Machine
Learning, Robotics, The Internet of Things, Neural Networks, Deep Learning, Rein-
forcement Learning and Our Future. Bravex Publications, 2019, 112 p.
11. M. Zgurovsky and Yu. Zaychenko, Big Data: Conceptual Analysis and Applications.
Springer Nature Switzerland AG, 2019, 306 p.
Received 13.06.2022
INFORMATIONON THE ARTICLE
Serhii V. Melnykov, ORCID: 0000-0003-2873-5730, International Scientific and Educa-
tional Center of Information Technologies and Systems, Ukraine, e-mail: psmail@i.ua
Petro M. Malezhik, ORCID: 0000-0001-6816-988X, Dragomanov National Pedagogical
University, Ukraine, e-mail: p.m.malezhyk@npu.edu.ua
Aydin С. Gasanov, ORCID: 0000-0002-5821-0751, Dragomanov National Pedagogical
University, Ukraine, e-mail: 0677937631@ukr.net
Petro I. Bidyuk, ORCID 0000-0002-7421-3565, Educational and Research Institute for
Applied System Analysis of the National Technical University of Ukraine “Igor Sikorsky
Kyiv Polytechnic Institute”, Ukraine, e-mail: pbidyuk_00@ukr.net
МЕТОДОЛОГІЧНІ АСПЕКТИ ІНТЕЛЕКТУАЛІЗАЦІЇ СИСТЕМ ОПЕРА-
ТИВНОГО КЕРУВАННЯ ДИНАМІЧНИМИ ОБ'ЄКТАМИ / С.В. Мельников,
П.М. Малежик, А.С. Гасанов, П.І. Бідюк
Анотація. Розглянуто питання організації інтелектуальних систем керування:
визначення рівнів інтелектуалізації, послідовність проведення аналізу складо-
вих компонентів процесу керування, унесення в систему характеристичних
властивостей інтелекту, особливості структурної організації, оцінювання
міри інтелектуалізації, приклади практичної реалізації. Метою розгляду є
висвітлення напрямів організації інтелектуальних систем керування,
необхідності і корисності системного розгляду, що враховує вимоги по-
ставленого цільового завдання, характеристики зовнішнього середовища,
використовувані засоби отримання, перетворення інформації і відпрацювання
керування, функціональні характеристики і досвід роботи людини-оператора.
У результаті проведеного аналізу виділено рівні розвитку інтелекту системи
керування, визначено етапи проведення інтелектуалізації системи, показано
ефективність викладених представлень для вирішення практичних завдань.
Ключові слова: система керування, інтелектуалізація, організація систем ке-
рування, інтелектуальне керування, людино-машинні системи.
|
| id | journaliasakpiua-article-275020 |
| institution | System research and information technologies |
| keywords_txt_mv | keywords |
| language | English |
| last_indexed | 2025-07-17T10:28:05Z |
| publishDate | 2022 |
| publisher | The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" |
| record_format | ojs |
| resource_txt_mv | journaliasakpiua/3b/4b62fc64a4619541071a30695a0b4c3b.pdf |
| spelling | journaliasakpiua-article-2750202023-05-21T20:04:38Z Methodological aspects of operative control system intellectualization for dynamic objects Методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами Melnykov, Serhii Malezhik, Petro Gasanov, Aydin Bidyuk, Petro система керування інтелектуалізація організація систем керування інтелектуальне керування людино-машинні системи control system intellectualization organization of control systems intelligent control man-machine systems The problems of intelligent control system organization are considered: determining the number of intellectualization levels, the sequence of actions required for analysis of the control process, adding to the control system new elements providing for enhancement degree of its intellectualization, special features of its structural organization, estimating the possibilities of intellectualization, providing examples of practical intellectualization. The primary purpose of the study is to determine the purposeful organization of intelligent control systems as well as the necessity and usefulness of systemic consideration that takes into consideration the following: requirements of the problem statement, characteristics of the environment, means for acquiring and processing necessary information, working control mechanisms, functional characteristics and experience of user-operator. As a result of the analysis performed, characteristic levels of the intellectual development of a system were determined, the stages of performing intellectualization of a control system were proposed, and the effectiveness of proposed solutions for practical problems was shown. Розглянуто питання організації інтелектуальних систем керування: визначення рівнів інтелектуалізації, послідовність проведення аналізу складових компонентів процесу керування, унесення в систему характеристичних властивостей інтелекту, особливості структурної організації, оцінювання міри інтелектуалізації, приклади практичної реалізації. Метою розгляду є висвітлення напрямів організації інтелектуальних систем керування, необхідності і корисності системного розгляду, що враховує вимоги поставленого цільового завдання, характеристики зовнішнього середовища, використовувані засоби отримання, перетворення інформації і відпрацювання керування, функціональні характеристики і досвід роботи людини-оператора. У результаті проведеного аналізу виділено рівні розвитку інтелекту системи керування, визначено етапи проведення інтелектуалізації системи, показано ефективність викладених представлень для вирішення практичних завдань. The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2022-12-27 Article Article application/pdf https://journal.iasa.kpi.ua/article/view/275020 10.20535/SRIT.2308-8893.2022.4.04 System research and information technologies; No. 4 (2022); 44-57 Системные исследования и информационные технологии; № 4 (2022); 44-57 Системні дослідження та інформаційні технології; № 4 (2022); 44-57 2308-8893 1681-6048 en https://journal.iasa.kpi.ua/article/view/275020/270176 |
| spellingShingle | система керування інтелектуалізація організація систем керування інтелектуальне керування людино-машинні системи Melnykov, Serhii Malezhik, Petro Gasanov, Aydin Bidyuk, Petro Методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами |
| title | Методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами |
| title_alt | Methodological aspects of operative control system intellectualization for dynamic objects |
| title_full | Методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами |
| title_fullStr | Методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами |
| title_full_unstemmed | Методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами |
| title_short | Методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами |
| title_sort | методологічні аспекти інтелектуалізації систем оперативного керування динамічними об'єктами |
| topic | система керування інтелектуалізація організація систем керування інтелектуальне керування людино-машинні системи |
| topic_facet | система керування інтелектуалізація організація систем керування інтелектуальне керування людино-машинні системи control system intellectualization organization of control systems intelligent control man-machine systems |
| url | https://journal.iasa.kpi.ua/article/view/275020 |
| work_keys_str_mv | AT melnykovserhii methodologicalaspectsofoperativecontrolsystemintellectualizationfordynamicobjects AT malezhikpetro methodologicalaspectsofoperativecontrolsystemintellectualizationfordynamicobjects AT gasanovaydin methodologicalaspectsofoperativecontrolsystemintellectualizationfordynamicobjects AT bidyukpetro methodologicalaspectsofoperativecontrolsystemintellectualizationfordynamicobjects AT melnykovserhii metodologíčníaspektiíntelektualízacíísistemoperativnogokeruvannâdinamíčnimiobêktami AT malezhikpetro metodologíčníaspektiíntelektualízacíísistemoperativnogokeruvannâdinamíčnimiobêktami AT gasanovaydin metodologíčníaspektiíntelektualízacíísistemoperativnogokeruvannâdinamíčnimiobêktami AT bidyukpetro metodologíčníaspektiíntelektualízacíísistemoperativnogokeruvannâdinamíčnimiobêktami |