Cybernetics, Engineering
Cybernetics, Engineering
a scientific field dealing with the use of standard cybernetics ideas and methods for the study of control systems. Engineering cybernetics is the scientific basis for the integrated automation of production processes and the development and construction of control systems in transportation, irrigation, and gas-distribution systems; atomic power plants; and spacecraft. The “man-machine” problem, which encompasses questions of rational distribution of functions between human beings and automatic devices in complex control systems (in which the human being participates directly as an essential link in the system), is one of the principal problems of engineering cybernetics.
The greatest integration of human and automatic functions is achieved in the cyborgs (“cybernetic organisms”), which are devices with a high degree of symbiosis in the physical and intellectual actions of the human being and the automatic equipment. Cyborgs, like manipulating robots, are becoming widely used to control objects under conditions of inaccessibility or danger to human life. Human participation in the functioning of automated control systems led to a situation in which the psychological state of the operator, in addition to his physiological traits, acquired considerable importance. Thus arose engineering psychology, a new direction in scientific research that is closely tied to engineering cybernetics; its most important task is the development of methods for using the psychophysiological characteristics of the human being in designing and operating complex man-machine control systems.
In solving many problems (for example, ship and aircraft navigation, the construction of measuring and monitoring instruments, and the development of automatic readers), specialists in engineering cybernetics try to apply to control technology methods developed by nature; this has led to the formation of a major independent area of study, bionics, which intersects engineering cybernetics.
Pattern recognition is one of the areas of investigation for engineering cybernetics. Recognition systems are used not only in the construction of reading machines but also for recognition and analysis of situations that characterize the state of production processes or physical experiments; the systems are also used in the development of automatic diagnostic equipment for medicine. Engineering cybernetics includes identification of control objects—that is, determination of the dynamic characteristics of the objects being controlled by observation and measurement of some of their parameters and external disturbing influences. The development and study of various methods of identification are independent areas of study in engineering cybernetics. Research in prediction theory and the development of automatic predicting machines may also be included in engineering cybernetics.
A characteristic feature of the development of engineering cybernetics in the late 1960’s and early 1970’s has been the extensive use of computer equipment in control systems, including automated control systems for enterprises. The construction of such systems is a complex and multifaceted task, which is based on engineering cybernetics, systems engineering, information theory, and economic cybernetics; it is not always possible to draw clear dividing lines among these scientific fields. Engineering cybernetics conducts research and solves problems related primarily to the lower levels of production control (the machine, the production process, and the shop system), whereas systems engineering concentrates on the middle levels (management of the enterprise, group of enterprises, or industry) and on automation of design processes and integrated research projects (for example, in geophysical and hydrophysical research). All levels of control are closely interwoven. Therefore, the building of an automated control system is approached as a single, integral problem, and the tasks of planning, development, manufacture, testing, troubleshooting, and operation are solved comprehensively. Purely technical aspects of the integral problem, as well as those of management, economy, sociology, law, and ethics, are taken into account. The construction of an automated control system for an enterprise requires extensive organizational and technical preparation. Organizational preparation involves, above all, algorithmic description of processes and the writing of algorithms for control of subsystems and of the system as a whole. Technical preparation involves the choice of standard hardware (computers, data display devices, and control consoles) required for effective functioning of the system and, when necessary, the development of new hardware.
Because of the saturation of control systems with various technical devices, automatic monitoring has gained in importance as a means of increasing their reliability of operation. As in the case of the general problem of increasing the effectiveness of the system, the solution of this problem is significantly tied to giving the human operator the necessary summarized visual information. Various means of information display (character indicators, mnemonics, lighted panel indicators, closed-circuit television, and special screens based on optical electronics and holography) have been constructed for this purpose, taking into account the psychophysiological characteristics of the human being that enable him to participate in the control process.
In most industrial control systems the a priori information needed for optimum control is lacking, and the human operator must accumulate it during the operation of the system. Therefore, the various adaptive systems that have been studied in the theory of automatic control are equally important in development of automatic control systems. This reveals the continuity —and even a certain coincidence—of the problems of the theory of automatic control and engineering cybernetics. This is also true of the study of the dynamic properties of automated control systems for enterprises (stability, control accuracy, and so on)— that is, the problem areas that determine the scientific content of both engineering cybernetics and the theory of automatic control.
The presence of a human being in a control system required the solution of many new problems that had not arisen during study of automatic control systems. In particular, it became necessary to study human intellectual activity during the process of control (logical description of the operator’s functioning and methods of describing purposeful behavior and the learning process). Because of the multiplicity of problems that arose in the study of man-machine control systems, it became necessary to find overall methods of research that encompassed many problems from a single point of view. In the 1970’s, therefore, engineering cybernetics began to develop in the direction of constructing and studying abstract models of complex control systems.
Methods of solving the problems of stability, optimality, and pattern recognition, as well as the study of finite automata and the solution of mathematical-economic problems—whose chief difficulty lies in the presence of a very large number of interacting elements (subsystems) included in the complex system—are taking on great importance in engineering cybernetics. The principal ways of surmounting these difficulties are the methods of decomposition and aggregation. Another important problem in engineering cybernetics is that of multiple criteria, which consists in choosing values for control actions that ensure that any optimum decision found for any subsystem will also be optimum (or suboptimum) for the system as a whole. Analytic methods of studying complex systems are very important for the study of real control systems for production and transportation, but at present they cannot be applied in practice because the problems are too complex; methods of simulation are more common for studying complex technical systems (as of 1972). In the simulation of man-machine systems, special simulation complexes— and even simulation centers—are built, whereas the study of automatic control systems requires only common methods of simulation (analog, digital, and hybrid, or analog-digital, methods). In addition to analog and digital computers, the special simulation facilities include various data display devices, specialized consoles, means of communication, and other equipment that makes possible the creation for the human operator of working conditions that are as close as possible to actual conditions.
REFERENCES
Ivakhnenko, A. G. Tekhnicheskaia kibernetika. Kiev, 1962.Teoriia avtomaticheskogo regulirovaniia, books 1–3. Moscow, 1967–69.
Tekhnicheskaia kibernetika v SSSR. Moscow, 1968.
Kibernetika i vychislitel’naia tekhnika. Fase. 1 :Slozhnye sistemy upravleniia. Kiev, 1969.
Voronov, A. A. Osnovy teorii avtomaticheskogo upravleniia, part 3. Moscow-Leningrad, 1970.
Tsien, H. S. Tekhnicheskaia kibernetika. Moscow, 1956. (Translated from English.)
Obshchaia teoriia sistem. Moscow, 1966. (Translated from English.)
Tekhnicheskaia kibernetika za rubezhom. Moscow, 1968. (Translated from English.)
Issledovaniia po obshchei teorii sistem. Moscow, 1969.
A. I. KUKHTENKO