ISSN 1991-2927
 

ACP № 3 (57) 2019

'Automation of Control Processes' # 2 (52) 2018

Contents
AUTOMATED CONTROL SYSTEMS

Dmitrii Vladimirovich Samoilenko, Mozhaisky Military Space Academy, Candidate of Engineering; graduated from Shtemenko Krasnodar Highest Military School; Postdoctoral Student of Mozhaisky Military Space Academy; an author of articles, inventions in the field of information security, crypto code information security systems, modular arithmetic of multidimensional numerical systems. [e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it. ]D. Samoilenko

Increasing of information survivability of the group of robotic engineering complexes under destructive attack of the violator52_1.pdf

A group of robotic engineering complexes such as unmanned aerial vehicles (UAVs) operating under an intelligent jamming attack of a violator is considered. In order to meet these operating conditions, a subsystem of cryptographic code protection of information based on the aggregated application of block encryption algorithms and polynomial noise-immune coding methods has been proposed. Further investigation disclosed that providing the subsystem under consideration with new properties ensuring the restoration of true encrypted information that is subjected to the destructive impact of the violator i.e. imitation proof as well as a regenerative process activation ensuring the integrity (restoration) of data i.e. information survivability are subsystem features. For the case of physical delivery of the information extracted by UAV complexes, it is suggested that the ensemble of storage devices onboard various UAVs with common functions be considered as a single system of storage devices with a subsystem of cryptographic code transformation of information. In this case, even the physical loss of the i-th UAV grouping as a storage node does not lead to complete or partial loss of data (including distorted data) and allows initiating the procedure for their recovery.

Group of unmanned aerial vehicles, cryptography, noise-immune coding, imitation proof, information survivability, information integrity.

Aleksei Aleksandrovich Gudkov, the Marshal Budjonny Military Academy of Signal Corps, Master’s Degree; graduated from the Cherepovets Military Engineering Institute of Radioelectronics; the Marshal Budjonny Military Academy of Signal Corps; Adjunct of the Marshal Budjonny Military Academy of Signal Corps; an author of proceedings and publications in the field of an assessment of electromagnetic accessibility and calculation of structural stability of hierarchical systems. [e-mail: gudkov_aa@rambler.ru]A. Gudkov,

Sergei Romanovich Malyshev, the Marshal Budjonny Military Academy of Signal Corps, Candidate of Engineering, Associate Professor; graduated from Mozhaisky Military Engineering Institute; Honored Inventor of the Russian Federation; Associate Professor of theMilitary Academy of communications an author of tutorials, articles and inventions in the field of modern theoretical aspects of radiomonitoring and radioengineering control. [e-mail: malishevsr56@ya.ru]S. Malyshev,

Sergei Vasilevich Krasnov, Graduate School of Business Management Technologies of St. Petersburg Polytechnic University, Candidate of Engineering, Associate Professor; graduated from the Ulyanovsk Higher Military Command School of Communications; an adjunct of Ulyanovsk Military Engineering Communication School; Associate Professor of Graduate School of Business Management Technologies of St. Petersburg Polytechnic University; an author of more than 50 scientific works in the field of design and implementation of information systems and technologies. [e-mail: hsm.krasnov@gmail.com]S. Krasnov

Parametric Synthesis of Radiomonitoring and Radiotechnical Control Systems Based on the Formal Approach 52_2.pdf

The task of synthesizing the radiomonitoring systems in the theory of complex systems is the task of estimating the optimality of the system or the problem of optimizing the system. The solution of the synthesis problem consists in considering a number of optimization problems, each of which characterized by the use of various optimality criteria. The diversity of radio electronic systems and the specific features of their functioning do not allow forming a unified methodology that allows the synthesis of any radiomonitoring system. In this regard, the synthesis of a particular system is an individual process. The concept of the synthesis of radiomonitoring system based on the general principles of the system approach in the synthesis of a complex system of any purpose, and, at the same time, includes a number of distinctive features due to the features of the system synthesized. The synthesis of the systems under consideration should include two stages: structural and parametric synthesis. At the first stage, indicators of results, requirements for results, criteria for assessing the quality of results are set. The second stage determines the efficiency index, the requirements for the level of efficiency, the criterion for assessing the effectiveness.

Parametric synthesis, optimization, radiomonitoring system.

Iurii Petrovich Egorov, Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’, Doctor of Engineering, Professor; graduated from the Faculty of Radioengineering at the Admiral Makarov Leningrad Higher Maritime School of Engineering; Chief Staff Scientist at Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’; specializes in the field of macro-design of large-scale information management systems; an author of monographs, articles, inventions in the field of computer-aided control systems design. [e-mail: yupe@mail.ru]I. Egorov,

Anatolii Ivanovich Piatakov, Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’, Candidate of Engineering; graduated from the Marshal Budjonny Military Academy of Signal Corps; completed his postgraduate studies ibidem; Chief Specialist of FRPC JSC ‘RPA ‘Mars’; specializes in the field of in organization and creation of discrete messages transmitting systems; an author of articles in the field of the automation complexes reliability and data communication. [e-mail: uljanovsk-anatol@mail.ru]A. Piatakov,

Lilia Irfanovna Suleimanova, Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’, Candidate of Engineering; graduated from the Power Engineering Faculty of Ulyanovsk State Technical University; Lead Engineer at FRPC JSC ‘RPA ‘Mars’; an author of articles and inventions in the field of heavy DC detector means. [e-mail: suleimanova.lili@mail.ru]L. Suleimanova

Estimation of Hardware and Software Complex Readiness to Solve the Functional Tasks 52_3.pdf

N view of the increasing complexity of performed hardware and software functions of automated management systems, designers focus on the task to determinate the probability of hardware and software availability for a time intended for solving the functional objective. The article deals with the problems of reliability prediction and estimation of hardware and software readiness to solve the tasks. Authors offer a model of hardware and software presented as a functional part set. The things that improving the accuracy of estimation of the hardware and software readiness to solve functional tasks are justified. Authors present an operational availability factor chosen as a key characteristic of the hardware and software readiness to solve functional tasks. It is calculated on the basis of both current and statistical reliability indices of system components that make the functional parts. A prediction algorithm for the loading of central processor unit, RAM and external memory of the computer system is proposed. It uses an effect of exponential smoothing of statistical data before the period of prediction. The recommendations for operators and operating personnel on the practical application of methods proposed are presented in the article.

Reliability, reliability evaluation, functional reliability, software and hardware reliability, reliability performances.

INFORMATION SYSTEMS

Aleksandr Alekseevich Emelianov, Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’, Candidate of Engineering; graduated from F.E. Dzerzhinsky Military Academy; Deputy Chief Engineer for Quality Assurance and Engineering Support - Head of the Management Department of Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’; an author of publications in the field of creation of the quality management and information security systems, statistical analysis of supplier appraisal. [e-mail: mars@mv.ru]A. Emelianov,

Iuliia Aleksandrovna Radionova, Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’, Candidate of Engineering; graduated from the Faculty of Mathematics and Mechanics of Ulyanovsk State University; finished her postgraduate study at Ulyanovsk State Technical University; Lead Software Engineer at FRPC JSC ‘RPA ‘Mars’; an author of articles in the field of automated workflow systems, intelligent technical documentation storages, and statistical analysis of supplier appraisal. [e-mail: julia-owl@mail.ru]I. Radionova,

Aleksandr Leonidovich Savkin, Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’, Candidate of Military Sciences, Associate Professor; graduated from Ulyanovsk Higher Military Command School of Communications, Marshal Budjonny Military Academy of Signal Corps, completed postgraduate studies in the Military Academy of Communications; Head of Science and Engineering Support Department of FRPC JSC ‘RPA ‘Mars’; an author of scientific works, textbooks, and articles in the field of development and modelling of communication control systems and statistical analysis of supplier appraisal. [e-mail: mars@mv.ru]A. Savkin

Information and Analytical Model of Product Testing for Counterfeit 52_4.pdf

The article considers a problem of counterfeit products and the existing methods for solving this problem. An information and analytical method for the optimization of an expert product evaluation is proposed, which uses a parameter set and allows to automate the final decision-making process. The set of parameters takes into account not only indicators for suppliers, but also possible consequences of using counterfeit products for different groups of consumers. The method is based on the application of a neural network using a certain base of expert knowledge. Network training is provided during analyzing process with the possibility of correction by an expert. Modeling of the analysis process with different sets of input parameters and various parameters of the neural network was carried out. The results are stored in the database; the neural network performance is evaluated based on these results. The database structure, input parameters structure and the visualization of experimental results are presented.

Counterfeit, modeling, evaluation parameter, database, software, neural network.

Aleksei Vladimirovich Neizhmak, he Marshal Budjonny Military Academy of Signal Corps, raduated from the Marshal Budjonny Military Academy of Signal Corps; Postgraduate Student in the Military Academy of Communications; an author of articles in the field of noise-immune coding. [e-mail: a.v.neizhmak@mail.ru]A. Neizhmak

The Model for the Operation of a Reliability Enhancement System Using Control Packages 52_5.pdf

The article deals with a formal description of the algorithm (process) of unidirectional transmission of information in a radio link with its feedback (unidirectional half-duplex). It does not violate the generality of the examination, since the process of simultaneous counterpropagating information in a radio link (duplex) is reduced to an independent description of two mutually independent unidirectional transmission processes. A description of the algorithm for the operation of a reliability enhancement system using control packages in terms of Markovian (semi-Markov) processes and the theory of stochastic control without considering the effect of electronic suppression are also given. For this purpose, the working states of the calling and answering subscribers were formulated and described, and a list of potential, allowed and reduced states was defined. For the purpose of explaining the principles of constructing the model, a radio link that functioned without taking into account the influence of the electronic suppression system was considered.

Reliability enhancement system, receipt, control packages, algorithm, transition state graph of a random process, calling subscriber, answering subscriber.

Aleksandr Nikolaevich Afanasev, Ulyanovsk State Technical University, Doctor of Engineering, Professor; graduated from the Radioengineering Faculty of Ulyanovsk Polytechnic Institute; First Vice-Rector, Vice-Rector for Distance and Further Education of Ulyanovsk State Technical University; an author of more than 200 articles in the field of CAD; research interests are computer-aided training system, the organization of computing processes and structures of computers, design of intelligent systems, CAD systems, management of complex workflows, diagrammatics of graphic languages. [e-mail: a.afanasev@ulstu.ru]A. Afanasev,

Sergei Igorevich Brigadnov, Ulyanovsk State Technical University, graduated from the Faculty of Information Systems and Technologies of Ulyanovsk State Technical University; Postgraduate Student of the Department of Computer Engineering of Ulyanovsk State Technical University; an author of more than 20 articles in the field of CAD; research interests are intelligent systems for the development of complex automated systems, automated learning environments. [e-mail: sergbrig@yandex.ru]S. Brigadnov

Methods and Means of a Complex System for the Analysis of Design Solutions and Designer Training 52_6.pdf

Currently, computer-aided design (CAD) systems have become an integral part of the work of designers, planners and architects. Modern CAD software systems allow to produce project documentation, perform engineering calculations, to perform modeling and rendering in 3D. A complex system of analysis of design solutions made in CAD KOMPAS-3D and the training of a designer is developed. The mathematical support of the integrated system for the analysis of design solutions and the training of the designer is developed: the method of structural and parametric analysis of project solutions, the associative-oriented model of the designer's competencies, the algorithm for forming the trajectory of training in computer-aided design. The architecture of the complex system is developed; its main components are described: the block of the analysis of design solutions, the block of training of the designer. The developed system supports the functions of automatic restructuring and classification of three-dimensional engineering products, as well as the formation of recommendations to the designer and personalized trajectory of learning computer-aided design in CAD KOMPAS-3D.

Cad kompas, analysis of design solutions, training of the designer, classification of products.

MATHEMATICAL MODELING

Innokentiy Vasilyevich Semushin, Ulyanovsk State University, Doctor of Science in Engineering, Professor of Information Technology at Ulyanovsk State University. Memberships in Professional Organizations: IEEE Society; IEEE Control Systems Society; “Russian Professorial Assembly”. Author of papers, monographs and textbooks; holds patents for inventions. Field of Interest: systems theory, control. [e-mail: kentvsem@yandex.ru]I. Semushin

Towards Robust Riccati Iterations for Lqg-optimal or Parameter-adaptive Estimation and Control Processes 52_7.pdf

The paper aims at the development of robust and efficient computational algorithms for stochastic linear control based on scalarized square-root implementations. In the capacity of starting point, it uses the classical (formal) solution to the control problem known from the three-decker monography Stochastic models, estimation, and control, Academic Press, 1979-1982, by Peter S. Maybeck. The mutually time-inverse Riccati iterations being a core part of the solution, are interpreted in a uniform notation as the two-stage update processes. For them, a transition to the two kinds of scalarized algorithms - direct and inverse, is performed to introduce into consideration a scalarized filter and scalarized regulator and avail ourselves of the opportunity to go into the question of numerically stable square-root computation designs for Riccati iterations. The paper demonstrates one possible configuration in the form of the Potter-style algorithm. That establishes the new direction in constructing robust LQG-regulators for control as being based on many advances in robust filtering.

Lqg-control, square-root algorithms, finite-horizon discrete-time control, scalarization.

Viacheslav Andreevich Sergeev, Ulyanovsk Branch of the Kotel’nikov Institute of Radioengineering and Electronics of the Russian Academy of sciences, Doctor of Engineering, Associate Professor; graduated from the Faculty of Physics of Gorky State Technical University; Head of Ulyanovsk Branch of the Kotel’nikov Institute of Radioengineering and Electronics (IRE) of the Russian Academy of sciences (RAS); Head of the Department of Radioengineering, Opto and Nanolectronics of Ulyanovsk State Technical University at Ulyanovsk Branch of the Kotel’nikov IRE; an author of monographs, articles, and inventions in the field of researching and simulating semiconductor device and integrated circuit performance, and measuring their thermal characteristics. [e-mail: sva@ulstu.ru]V. Sergeev,

Sergei Evgenevich Rezchikov, Ulyanovsk Branch of the Kotel’nikov Institute of Radioengineering and Electronics of the Russian Academy of sciences, graduated from the Faculty of Radioengineering of Ulyanovsk State Technical University; Postgraduate Student of the Base Department of Radioengineering, Opto and Nanolectronics at the Kotel’nikov Institute of Radioengineering and Electronics (IRE) of RAS; an author of articles in the field of measurement automation and noise characteristics research of semiconductor devices and integrated circuits. [e-mail: s.rezchikov@ulstu.ru]S. Rezchikov

The Optimization of Procedures for Measuring the Semiconducter Devices’ Paramters of Low-freqency Noise Influenced By White Noise 52_8.pdf

The brief analysis of techniques for measuring the parameters of low-frequency (LF) noise with 1/f?-type spectrum has been proposed. The estimation of measurement error of power spectral density (PSD) and ?-rays spectrum index are given by taking into account the influence of white-noise level. Based on the results of noise spectral power density measurements implemented at three frequencies when series and parallel filtering, authors propose procedures of measuring the index of ?-rays spectrum that minimize a total inaccuracy of ? value determination. The essence of optimization in case of serial filtering reduces to the optimum distribution of preset overall measurement time between measurements at the first, second and additional frequencies in case of the given frequency ratio, and when parallel filtering, to the determination of the optimum frequency ratio of measuring the noise PSD in case of the preset overall measurement time. A systematic measurement error of ? value has been estimated while implementing of various measurement procedures depending on ? value. The estimations are presented in the article. It is shown that the optimization of measuring procedures allows to reduce an error of the ? value determination by 1.5 to 2 times.

Low-frequency noise, power spectral density, index of spectrum form, measurement, error, optimum procedures.

Iurii Dmitrievich Ukraintsev, Ulyanovsk State University, Candidate of Engineering; graduated from the Faculty of Radiocommunications of the Leningrad Military Academy of Communications; Associate Professor of the Department of Telecommunication Technologies and Networks at Ulyanovsk State University; an author of articles, monographs inventions in the field of statistical analysis and synthesis of telecommunication systems. [e-mail: 9019471930@mail.ru]I. Ukraintsev,

Vladimir Viktorovich Kalnikov, Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’, Candidate of Engineering, Associate Professor; graduated from Radioengineering Faculty of Kiev Higher Military Engineering College of Communications; Chief Specialist at FRPC JSC ‘RPA ‘Mars’; specializes in the field of special-purpose control system design, creation of communication and data exchange systems; an author of articles and inventions in the field of designing the distributed control systems of special-purpose, communication and data exchange systems. [e-mail: mars@mv.ru]V. Kalnikov

An Alternate Bayesian Approach to the Detection of an Adaptive Decision Threshold Scheme for the Receiver At Decameter Radio Links 52_9.pdf

The article deals with an approach to the determination of the decision threshold scheme of a receiving device in decameter radio link, which implements the alternative Bayesian method. In contrast to the classical approach when the decision threshold scheme is determined with the assumption that the probability density function (PDF) of the received signal is subject to the Gaussian low, an approach based on the current recovery of instantaneous values of the received signal envelope in each distribution path is proposed. The PDF recovery algorithm is based on the nonparametric Parzen-Rosenblatt density estimation method. The estimating is carried out during a specified communication session. In this case, the most probable values of the signal (mode) are determined if the noise interference is absent and present, the decision threshold scheme is established in accordance with the real situation on the communication line. Then, on the basis of the mode shift, a decision is made about the signal-error probability. On the basis of simulation modeling, the efficiency of the approach is shown when operating at decameter radio links working under nonparametric prior uncertainty relative to signals and interference.

Signal depression, instantaneous values of the received signal envelope, parzen method of the probability density estimation, mode, decision threshold scheme of a receiving device, error probability.

Irina Aleksandrovna Sedykh, Lipetsk State Technical University, Candidate of Physics and Mathematics; graduated from the Faculty of Automation and Informatics of Lipetsk State Technical University (LSTU); Associate Professor of the Mathematics Department at Lipetsk State Technical University; an author of monographs and articles, owner of certificates of registration of the computer programs in the field of neighborhood modeling of dynamic systems. [e-mail: sedykh-irina@yandex.ru]I. Sedykh,

Evgenii Sergeevich Anikeev, Lipetsk State Technical University, graduated from the Faculty of Physics and Engineering of Lipetsk State Technical University (LSTU); Master’s Degree Student at Lipetsk State Technical University; an author of articles, owner of certificates of registration of the computer programs in the field of neighborhood modeling of dynamic systems. [e-mail: evgenij-anikeev@yandex.ru]E. Anikeev

Hierarchical Colored Time Petri Nets on the Basis of Neighborhood Models 52_10.pdf

The article describes the definition and algorithm for the operation of common labeled Petri nets. Some varieties of Petri nets such as timed, colored and colored timed Petri nets are shown. The proposed hierarchical colored Petri nets based on neighborhood models are an extension of the considered classes. The hierarchy in this system is represented by a set of positions and transitions, in each of which a neighborhood model can be embedded. The concept of a dynamic neighborhood model is also given in this paper. Adding a hierarchy allows to analyze additional properties of simulated dynamic distributed processes. For colored Petri nets with hierarchical transitions, an operation algorithm is formulated that is designed and implemented as a Java program. The peculiarity of the models under consideration is the synchronization of the operating time of the external and internal systems. In addition, the control signals of the embedded neighborhood model depend on the change in the current marking of the Petri net when a hierarchical transition occurs. An example of the functioning of a Petri net with a hierarchical transition is given.

Modelling, colored timed petri net, hierarchical petri net, algorithm, neighborhood model.

Dmitrii Anatolevich Zhukov, Ulyanovsk State Technical University, graduated from the Faculty of Information System and Technologies of Ulyanovsk State Technical University; Postgraduate Student of the Department of Applied Mathematics and Computer Science; an author of proceedings in the field of the statistical methods and machine learning. [e-mail: zh.dimka17@mail.ru]D. Zhukov,

Vladimir Nikolaevich Kliachkin, Ulyanovsk State Technical University, Doctor of Engineering; graduated from the Mechanical Faculty of Ulyanovsk Polytechnic Institute; Professor at the Department of Applied Mathematics and Informatics of Ulyanovsk State Technical University; an author of scientific papers in the field of reliability issues and statistical methods. [e-mail: v_kl@mail.ru]V. Kliachkin

The Effect of the Control Sample Volume on the Quality of Diagnostics of the Technical Object State 52_11.pdf

He problem of predicting the serviceability of a technical object in terms of its performance is considered. The source data are the known results of the evaluation of the state of an object based on the results of the previous operation: If the specified values of controlled indicators technical system intact or defective. Such a problem can be solved by methods of machine learning, it reduces to a binary classification of the states of the object. The quality of diagnostics can significantly depend on many factors: the method of training, the correct allocation of factors characterizing the operation of the object, the volume of the sample, and others. The work studies the effect of the control sample volume on the quality of diagnosis, estimated by the number of mis-predicted states using the cross-validation method. The tests were carried out in the Matlab package, ten different training methods were used: logistic regression, support vector method, decision tree bugging, and others. It is shown that the correct choice of the proportion of the control sample can improve the diagnostic quality to 5-7%.

Technical diagnostics, serviceability of the indicator of functioning, machine learning, control sample, cross-validation.

ARTIFICIAL INTELLIGENCE

Gennadii Pavlovich Vinogradov, Tver State Technical University, Doctor of Engineering; graduated from Kalinin Polytechnic Institute; Professor of the Department of Informatics and Applied Mathematics of Tver State Technical University; an author of more than 200 scientific papers in the field of decision making theory, fuzzy logic, and hybrid systems. [e-mail: wgp272ng@mail.ru]G. Vinogradov

The Model for Decision Making Process in Situations With Incomplete and Fuzzy Information 52_12.pdf

It is shown that an agent when making decisions uses three sets of alternatives: control, structural, and identification. This presupposes the existence of three virtual parties involved in the selection of appropriate alternatives. Rules for the selection of alternatives depending on the subject's understanding of the situation and structure their interests are shaped by compromise. The work seeks to investigate the causes of discrepancies between actual and "optimal" choice. To this end, we propose a formal scheme included in the choice model of the stages of cognition and behavior associated with the choice. The proposed approach is based on a formalization of the ideas of Miller, and Pribram Galanter and uses the ideas of subjectively rational choice developed based on the theory of fuzzy sets. Subjectively rational choice assumes that the motivation of the choice is determined by both external and internal factors. Internal factors reflect the interests of the subject determined by its needs and ethical system to which it adheres. Evaluation of satisfaction with the current situation purposeful state entity, as shown, can lead to changes in the structure of interests of the subject, and it can choose.

Reflexive governance, decision-making model, a compromise.

Sergei Vasilevich Voronov, Ulyanovsk State Technical University, Candidate of Engineering; graduated from the Faculty of Radioengineering of Ulyanovsk State Technical University; Associate Professor of the Radioengineering Department of UlSTU; an author of articles, monographs in the field of digital signal and image processing and computer vision. [e-mail: valmedia@yandex.ru]S. Voronov,

Rinat Nailevich Mukhometzianov, University of Waterloo, Canada, Postgraduate Student at the University of Waterloo, Canada; graduated from the Faculty of Radioengineering of Ulyanovsk State Technical University; an author of articles in the field of computer vision. [e-mail: mukhometzyanov@mail.ru]R. Mukhometzianov,

Ilia Vasilevich Voronov, Ulyanovsk State Technical University, Postgraduate Student at Ulyanovsk State Technical University; graduated from the Faculty of Radioengineering of Ulyanovsk State Technical University; an author of articles in the field of digital signal and image processing. [e-mail: ilvo1987@gmail.com]I. Voronov,

Vadim Andreevich Shramov, Federal Research-and-Production Center Joint Stock Company ‘Research-and-Production Association ‘Mars’, Postgraduate Student at Ulyanovsk State Technical University; graduated from the Faculty of Radioengineering of Ulyanovsk State Technical University; Research-Engineer at FRPC JSC ‘RPA ‘Mars’. [e-mail: vadim_shramov@mail.ru]V. Shramov

Trafic Sign Detection and Recognition in Real Time on Mobile Devices 52_13.pdf

An automatic traffic sign recognition system localizes road signs from images captured by an onboard camera of a vehicle and determine what road signs depicted. Such systems may support the driver on the road, be part of the self-driving cars or advanced driver assistance systems. This paper proposes an approach for solving traffic sign recognition problem using deep learning methods adopted to mobile devices with low power consumption. The approach consists of two consecutive stages: detection of a traffic sign and recognition of a class of the detected sign. Data for analysis were taken from three open sets of images. In order to analyze the effectiveness of the solution obtained, the results achieved were compared with the results of well-known approaches to object detection based on the use of deep convolutional neural networks. The results showed that the proposed algorithm provides the best recognition quality for all used data sets, as well as the highest recognition rate.

Deep learning, viola-jones detector, convolutional neural networks, traffic sign recognition, object detection.

Amir Muratovich Akhmetvaleev, Kazan National Research Technical University named after A.N. Tupolev-KAI, Postgraduate Student of Kazan National Research Technical University named after A.N. Tupolev-KAI; graduated from the Faculty of Technical Cybernetics and Informatics of the Kazan State Technical University named after A.N. Tupolev-KAI; an author of scientific works in the field of mathematical modelling, data analysis and machine learning methods. [e-mail: amir.akhmetvaleev@gmail.com]A. Akhmetvaleev,

Aleksei Sergeevich Katasev, Kazan National Research Technical University named after A.N. Tupolev-KAI, Candidate of Engineering; graduated from the Faculty of Physics and Mathematics of Elabuga State Pedagogical Institute; Associate Professor of the Information Security Systems Department of Kazan National Research Technical University named after A.N. Tupolev-KAI; an author of scientific works in the field of mathematical modelling, data analysis and development of intelligent decision support systems. [e-mail: Kat_726@mail.ru]A. Katasev

Instrumental Software Complex for Automation to Determine the Functional State of Intoxication of a Person 52_14.pdf

This article actualizes the need to determine the functional state of a person. Its decision is based on the application of the pupillometry method, which allows to determine the state of a person by his pupillary reaction to light changes. In order to automate the processes of determining the functional state of a person, an instrumental program complex is developed, based on the use of the neural network model. Its structure, composition and characteristics of components are described. The functioning of the program complex is considered on the example of the modules of neural network model construction, estimation of its accuracy on the basis of bootstrapping method, structural optimization of the model, and determining the functional state of a person. A number of researches and experiments were conducted on the basis of the program complex. The results of the influence of the number of stages of bootstrapping on the accuracy of the neural network model, the results of reduction of neural networks, the comparison of the accuracy of the model with the accuracy of other classification methods are presented. The results of the research showed the effectiveness of the neural network model and the possibility of its practical use to determine the functional state of the person in various subject areas.

Human functional state, papillomometry, neural network model, genetic algorithm, bootstrapping.

© FRPC JSC 'RPA 'Mars', 2009-2018 The web-site runs on Joomla!