Search

The paper proposes a method of infrared spectroscopy for objective control of the adhesive composition in the production of paper. For sizing paper, in order to increase mechanical strength and reduce water absorption, water-repellent materials are used. One of the most effective and used water-repellent material is ordinary pine resin rosin which is in short supply for the Republic of Uzbekistan. The paper investigates the properties of a new adhesive based on rosin from cherry resin for use in mass sizing in order to expand the range of sizing agents and reduce the share of imported components in the structure of local paper production. Infrared spectroscopy was used to identify the rosin-based glue from the resin of the cherry tree. The built-in device with a diamond crystal of the universal and compact Thermo Scientific Nicolet iS50 IR Fourier spectrometer made it possible to register spectra in the mid and far IR ranges up to 100 cm–1. The efficiency of sizing paper samples was studied by standard methods and evaluated by the following properties: density, tensile strength at stretching on a tensile machine, surface absorbency with one-sided wetting. Infrared spectra revealed the identity of the sizing agent from the resin of the cherry tree to the glue from the pine resin, taken for comparison, because their chemical composition is the same. A comparative analysis of the mechanical and surface properties of paper samples confirmed the possibility of using the studied sizing agent, since the obtained values are close to each other, while it is advisable to add in the range from 1 to 1.5 g per 100 g of dry cellulose raw materials. The use of infrared spectroscopy for express analysis of the composition of the sizing agent is promising, since it becomes possible to control the technological process and create glued types of paper products with desired properties that ensure print quality without loss of fine image details.

Personalization is one of the keyways to improve the performance of conversational agents. It improves the quality of user interaction with a conversational agent and increases user satisfaction by increasing the consistency and specificity of responses. The dialogue with the agent becomes more consistent, the inconsistency of responses is reduced, and the responses become more specific and interesting. Training and testing personalized conversational agents requires specific datasets containing facts about a persona and texts of persona’s dialogues where replicas use those facts. There are several datasets in English and Chinese containing an average of five facts about a persona where the dialogues are composed by crowdsourcing users who repeatedly imitate different personas. This paper proposes a methodology for collecting an original dataset containing an extended set of facts about a persona and natural dialogues between personas. The new RuPersonaChat dataset is based on three different recording scenarios: an interview, a short conversation, and a long conversation. This is the first dataset for dialogue agent personalization collected which includes both natural dialogues and extended persona’s descriptions. Additionally, in the dataset, the persona’s replicas are annotated with the facts about the persona from which they are generated. The methodology for collecting an original corpus of test data proposed in this paper allows for testing language models for various tasks within the framework of personalized dialogue agent development. The collected dataset includes 139 dialogues and 2608 replicas. This dataset was used to test answer and question generation models and the best results were obtained using the Gpt3-large model (perplexity is equal to 15.7). The dataset can be used to test the personalized dialogue agents’ ability to talk about themselves to the interlocutor, to communicate with the interlocutor utilizing phatic speech and taking into account the extended context when communicating with the user.

In this paper, we study the trajectory tracking problem of a three-wheeled omnidirectional mobile robot with full state constraints and actuator saturation. Firstly, we analyze a three-wheeled omnidirectional mobile robot and give control model with actuator saturation. By using tan-type Barrier Lyapunov Function and backstepping method, kinematic and dynamic controllers are built, which can ensure that the system full states will not violate the given constraints when the robot is performing trajectory tracking. Then, considering the differential explosion problem which occurs when solving the derivatives of the virtual control law, we use a second-order differential sliding mode surface to calculate it, so as to reduce the complexity of the operation. In addition, due to the output saturation problem of the robot drive motor, an auxiliary compensation system is adopted to compensate for the error generated by the saturation function. Finally, an experimental simulation is performed in MATLAB and the simulation results illustrate the effectiveness of the control algorithm proposed in this paper.

The paper analyzes the features of representing artificial neural networks in Simulink and SimInTech. Examples of visual schemes (models) built in these modeling environments using neural network blocks are given. The following shortcomings of such representations are the lack of mechanisms: for carrying out structural optimization of neural networks, for combining them into ensembles, for training them synchronously with the simulation of the object model. It was noted that there are difficulties in using other tools, such as specialized Python libraries (Keras, PyTorch, etc.), the NeuroGenetic Optimizer (BioCompSystems) for building neural network control models. A method is shown to implement the representation of neural networks in the formalism of the method of multilevel component circuits, according to which the construction of models of an object and a control system is carried out in a visual language from ready-made blocks (components) with directional and non-directional connections. A technique has been developed for multilevel representation of neural network control models, which allows them to be combined with other tools of the component circuit method. Two options for representing neural networks are proposed: with an encapsulated structure and with a component structure. The first version of the representation is characterized by the compactness of the representation of the control model, the possibility of automated variation and optimization of the structure of the neural network, and the possibility of changing the structure of the network during the executing of the model within a computational experiment (scenario). The second option has the ability to perform detailed debugging and research of the network learning process, and the ability to construct a network of any structural complexity. The paper describes the main developed components with their connections: a neural network, a training block, an ensemble unit (bagging), a block for reading data from a file, a sampling block, a neural network layer (input, hidden, output). A multilevel computer model of the uncontrolled flight of a body (target) and the controlled flight of a projectile is presented as an example to illustrate the operation of the developed components to solve the problem of controlling a projectile to hit the target. The developed component libraries can be used as part of the MARS modeling environment to build multilevel control systems for objects of a multiphysics nature.

The paper proposes a modified method for measuring the refractive index of a triangular prism using an autocollimation angle measuring system designed to measure angles formed by flat surfaces of objects. The method involves using a fixed mirror to reflect the refracted beam, measurement of the angles of incidence of the beam on the face of the prism corresponding to the autocollimation positions and calculating the refractive index of the prism material using the solution of a system of equations. The paper presents the experimental study results for a triangular prism made of K8 optical glass using the proposed method, and a comparison of the results with the readings obtained using the State Primary Standard of the Refractive Index Unit GET 138-2021. The proposed method makes it possible to simplify the process of measuring the refractive index since there is no need to measure the angle of deviation of the beam.

A wide class of smooth continuous dynamic nonlinear systems (control objects) with a measurable state vector is considered. The problem of finding a special function (Lyapunov function), which guarantees asymptotic stability for the presented class of nonlinear systems in the framework of the second Lyapunov method, is posed. It is known that the search for the Lyapunov function is an extremely difficult problem that has no universal solution in stability theory. The methods of selection or search of the Lyapunov function for stability analysis of closed linear stationary systems and for nonlinear objects with explicitly expressed linear dynamical and nonlinear static parts are well studied. At the same time, no universal approaches to finding the Lyapunov function for a more general class of nonlinear systems have been identified. In this paper, we propose a new approach to the search of the Lyapunov function for analyzing the stability of smooth continuous dynamic nonlinear systems with a measurable state vector. The essence of the proposed approach consists in the representation of some function through the sum of nonlinear summands representing the elements of the object state vector multiplied by unknown coefficients. The search for these coefficients is performed using a classical genetic algorithm including mutation, selection, and crossover operations. The found coefficients provide all the necessary conditions for the Lyapunov function (within the framework of the second Lyapunov method). The genetic algorithm approach does not require a training sample which imposes restrictions in the form of the structure of control objects included in it. A new method for finding the Lyapunov function represented as a nonlinear series with known functions multiplied by unknown coefficients is proposed. The effectiveness of the proposed method is demonstrated using computer simulations with a fixed number of iterations and varying population size. The dependence of the number of successfully found Lyapunov functions on the number of iterations of the genetic algorithm has been established. The convergence of the genetic algorithm using Holland’s schemes is analyzed. It is shown that the values of the sought coefficients of the potential Lyapunov function, at each algorithm iteration, approach the coefficients of the Lyapunov function which was also represented as a Taylor series. The method proposed in this paper outperforms known analogs in terms of speed, considers the decomposition of the potential Lyapunov function into a Taylor series with unknown coefficients, instead of using counterexamples or template functions.

This paper studies the trajectory tracking problem and the controller gain adjustment problem for Wheeled Mobile Robots. The controller gain has a great influence on the robot’s trajectory tracking: it can influence both the tracking accuracy and the tracking speed. Therefore, it is very important to choose a suitable control gain during the controller design process. Current neural network gain controllers have a complex structure and require a lot of calculations to find the optimal value. To solve this problem, we design a trajectory tracking controller with a simple structure with adaptive gain by combining the controller with a neural network. The input to this controller is the robot’s attitude error. The controller has no hidden layer and directly outputs the trajectory tracking control law. Firstly, the kinematic controller is designed based on Lyapunov function method to ensure that the robot moves according to the reference trajectory. Then, the online gain adjustment algorithm is designed by using neural network to realize the fast adjustment of the controller gain and ensure the reliability of the controller. Finally, the backstepping method is utilized to design the velocity tracking controller based on the error between the virtual velocity and the actual velocity. Considering the influence of the external environment, we also design a nonlinear disturbance observer to estimate the total disturbance on the robot. We perform simulation experiment in MATLAB. The result of the experiment shows that the control algorithm proposed in this paper can realize the accurate tracking of the robot on the specified trajectory. The gain adjustment algorithm we designed can find the optimal gain value quickly and efficiently, thus improving the stability and efficiency of the controller. The method can be applied to most mobile robot trajectory tracking problems and solves the problem of control gain adjustment.

The progress in the stream of the power electronic converters has led to the expansion of various protection devices for the distribution system. This also has led to an assortment of flexible transmission devices aiming to enhance the stability of the system throughout a variety of power quality issues and, furthermore, for enabling flexible uninterrupted power transmission during turbulences. This paper augments the employment of two Custom Power Devices, namely, Dynamic Voltage Restorer and Distribution Static Compensator for dealing with various power quality issues associated with distributed generation systems. This paper also involves analysis of performance of proposed Custom Power Devices with various algorithms, like gravitational search algorithm, BAT algorithm and ANT colony optimization algorithm for improving the stability of the power system. The proposed system has been tested with various distributed systems, fault conditions, and assessment has been performed among different algorithms in terms of supply voltage, supply current, active power, reactive power, and power factor. The design and analysis of entire system has been executed using MATLAB/Simulink.

The introduction of new types of heat exchangers with phase transitions and the solution of problems of optimizing the design and operational characteristics are a priority within the framework of the energy saving program. Known methods for calculating the heat-hydrodynamic parameters of the flow of refrigerants often do not take into account the specifics of boiling processes at low temperatures as well as in channels with a small flow area. This paper presents the results of modeling heat transfer during the boiling of refrigerants in the channels of evaporators of heat and cold energy complexes, taking into account the true flow parameters. The proposed mathematical model of the boiling of the working substance in channels of various shapes is based on the true flow parameters which imply knowledge of the channel cross-sectional areas occupied by each of the phases. The value of the true volumetric steam content provides the most correct modeling of two-phase flows in a wide range of regime and geometric parameters. The paper uses the equations of material and heat balance in combination with the equation of heat transfer from the environment to the boiling refrigerant. The map of flow regimes is used as an empirical component. A program has been developed for calculating the proposed system of equations which is solved iteratively at each time step using the finite volume method. Comparison of calculation results with experimental data on models of round and rectangular channels with intracanal boiling of refrigerants at positive and negative saturation temperatures is performed. It is shown that the calculation error does not exceed 10 % for a round and 20 % for a rectangular flow section. The verification results showed the possibility of using the model in the framework of engineering calculations. The proposed mathematical model can be used as the basis for the calculation programs for existing evaporators and for the creation of new types of heat exchangers with in-tube boiling of the working substance. The proposed method allows optimizing both geometric and thermal-hydrodynamic parameters.

The article discusses the experimental and theoretical results of studying the optical properties for planar quartz surfaces modified with gold nanostars using Surface-Enhanced Raman Spectroscopy (SERS). Surfaces functionalized with nanoparticles, prototypes of optical sensors, have been obtained. It is shown that the obtained experimental Raman signal enhancement from the dye rhodamine 6G (R6G) can be at least order 104. The paper presents the results of calculation, synthesis and study of the optical and morphological properties of such structures. A complex method for creating and evaluating the optical properties of planar nanostructures and evaluating their optical properties is proposed. The method includes the approaches based on mathematical modeling using finite differences in the time domain (FDTD - Finite- Difference Time-Domain) evaluating electromagnetic field strengths near the surfaces of star-shaped gold nanoparticles in colloidal solutions and on a silicon surface. During the simulation, we studied such parameters as the particle size, the wavelength of the exciting radiation, and the dependence of the effective amplification of the electromagnetic field on the morphological parameters of the star. The theoretical approach to the calculation of such structures is shown to be promising for the subsequent chemical synthesis described in the article and the study of optical properties. In this paper, a theoretical calculation of the parameters of the electric field and optical properties is carried out near star-shaped nanoparticles of varying sizes. These parameters were calculated by the finite difference method in the time domain. The task included finding the optimal sizes of nanoparticles as well as varying the thickness of the surface layers in order to obtain the maximum scattering values and further use of similar structures in SERS experiments. Also, within the framework of this work, syntheses of star-shaped nanoparticles was carried out, their optical and morphological properties were studied, quartz surfaces were functionalized with nanoparticles, and sensor prototypes were created in order to further assess the Raman signal amplification. Based on the results of mathematical modeling, the optimal sizes for the synthesis of star-shaped NPs were determined. Synthesis of NPs was carried out by a chemical method using seeds. To carry out the experimental part, SERS spectra were obtained using Centaur U spectrometer (OOO NanoScanTechnology, Russia). The spectrometer was equipped with three different laser sources: 632.8 nm He-Ne laser (17 mW), 532 nm and 473 nm DPSS lasers (45 mW). During the experiment, a He-Ne laser with λ = 632 nm and DPSS laser with λ = 532 nm were used to match the excitation of the plasmon maximum for the nanostars. The optical scheme of the spectrometer included an Olympus BX41 microscope (Olympus, Japan) with a 100× objective (NA 0.9) for positioning the beam and collecting scattered photons. The monochromator of the spectrometer had a focal length of 800 mm. A holographic diffraction grating of 300 gr/mm, and was equipped with a 1024×256 pixel thermoelectric cooled CCD detector (Andor Tech., UK). As a result of the paper, surfaces functionalized by nanoparticles were obtained prototypes of optical sensors. It is shown that the experimental gain of the Raman signal can be no less than 104 times. The presented data will serve as the basis for obtaining the required sizes of nanostars in the methods of the controlled chemical synthesis of colloidal nanoparticles. The results presented in the article can be applied to the development of biocompatible and highly sensitive optical sensors based on the effect of Raman scattering of light.

Residual amplitude modulation in LiNbO3 phase modulator is one of the key factors that limit the accuracy of highsensitive fiber-optical sensors. A fiber-optical gyroscope is an angular velocity sensor whose sensitivity is better than 0.001 °/h. Optical light intensity changes after phase modulator is a reason for wrong phase difference that introduces an error in the angular velocity signal. Most existing residual amplitude modulation suppression methods are based on reduction of back reflections between an optical fiber and integrated optical waveguide, absorbing groove to suppress or reduce reflection on the bottom face, and algorithmic compensation. In this paper, new approach to reduce residual amplitude modulation in LiNbO3 for fiber optical gyroscope application is presented. Method feature is an application of additional differential signal modulation with uniform amplitude distribution in the input signal voltage range of the phase modulator. The proposed method allows to suppress residual amplitude modulation of the multifunctional integrated optical circuit phase modulator more than 3 times using additional triangle signal modulation with the frequency f = 200.09 MHz and power P = 36 dBm. This method is suitable for improving fiber optical gyroscope accuracy. Moreover, it could be applied for any fiber-optic sensors based on LiNbO3 phase modulator. The paper will be of interest to specialists in the field of highly sensitive fiber optical sensors, fiber, and integrated optics.

Monitoring systems of automated production lines requires high accuracy and processing speed of acquired data. Nowadays digital twin technologies are rapidly developing as part of Industry 4.0. Digital twin is a virtual representation of a physical system that mimics the behaviour of a real object, and is used in real-time control, monitoring and disturbance prediction, without influence on a real object. Digital twin is being used in many fields of application such as: healthcare, manufacturing, education, city development, etc. Nonetheless, despite the rising popularity of digital twin researches, there is no basic approach to digital twin synthesis. Mostly, in state of the art articles, systems with known parameters are considered. In this paper, approach to development of digital twin, based on internal model for multi-link mechanisms with unknown motion parameters, is presented. Described approach intended to ease the task of production line monitoring and extend digital twin technology application field. In this article adaptive controller design, based on internal model, is presented, theoretical explanations of design process are provided, and application of the control algorithm in the task of multi-link mechanism motion trajectory parameters estimation is described. In this work the experimental synthesis of digital twin for Kuka youBot manipulator based on presented approach in real time is described. Adaptive motion parameters estimation was made for two links of the manipulator for the sake of confirmation of presented approach usage in the task of digital twin for systems with unknown parameters. Motion trajectory parameters of the robot links were defined manually with the use of chaotic generators to estimate the accuracy of the adaptive control system. This approach was used in order to calculate the error of reference tracking by comparing the control signal for digital twin with the reference exogenous input for manipulator links. As the results of the experiment, the graphs of the error reference tracking for both links are presented. As the goal of the experiment, the convergence of reference tracking error to the field [–0.005, 0.005] radian was set, and as can be seen from graphs this goal was achieved. There is a small disturbance of the error convergence on the graphs within the desired field which was caused by noised measurements and delays of the chosen modelling environment real-time simulations. In the conclusion of this paper, further work to reduce the influence of mentioned causes is described. The results of this paper can be used in further digital twin technology researches. Presented approach can be implemented in robots remote control tasks, production lines technical condition monitoring and in general motion trajectory parameters estimation tasks for multi-link mechanisms.

This paper discusses the effect called undesirable extrusion dynamics that occur during material extrusion printing or fused deposition modeling. This effect is revealed during high acceleration and deceleration of printing head and results in over-extrusion on corners of parts and printed layer unevenness. To minimize this effect, modern control systems use Advance algorithms for dynamics control. However, such disadvantages as inertia, reduced printing speed, need for manual calibration, as well as uncertainty of the influence of material, nozzle geometry, and printing process parameters on the algorithm performance do not allow this group of solutions to be applied on an industrial scale. The paper presents a study of the influence of printing characteristics, such as material type, extruder type, printing temperature, layer and nozzle geometry, on extrusion dynamics through a series of experiments. The experiments were carried out on a Creality Ender 3 printer. The obtained experimental data will allow us to deepen the understanding of extrusion dynamics influence in FDM printing; they are also used in the present study to modify existing dynamics control algorithms. The paper proposes a modification of Advance algorithm based on machine learning. It is proposed to implement an algorithm containing two trained neural network models. One model predicts changes in printing head motion to minimize residual defects and increase average print speed. The second model predicts the compensation parameter for specific printing conditions without the need for manual calibration. A neural network model was trained to determine the compensation parameter depending on the type of material, layer thickness, nozzle geometry and printing temperature. The model was trained based on experimental data. The developed algorithm was introduced into Linear Advance algorithm of the Marlin firmware and tested on Creality Ender 3. The experiments showed that the developed model can successfully predict the required compensation and can be applied during the printing process. The proposed algorithm helps to facilitate and automate the process of extrusion dynamics compensation which will expand the possibilities of FDM printers’ application in industrial conditions. The obtained algorithm can improve the accuracy and printing speed which will subsequently help to increase the economic independence and competitiveness of small design organizations and enterprises in Russia that use 3D printers. This research expands the possibilities of rapid prototyping and may help to ensure the rapid creation of pilot batches.

The paper deals with the problem of acoustic events classification which is actively applied to the problems of a safe city, smart home, IoT devices, and for the detection of industrial accidence. A solution to improve the accuracy of classifiers without changing their structure and collecting additional data is proposed. The main data source for the experiments was the TUT Urban Acoustic Scenes 2018, Development Dataset. The paper presents the way to increase the accuracy of audio event classification by using the FN-corrector. The FN-corrector is a linear two-stage classifier performing the transformation of the feature space into a linearly separable space and the linear separation of one class from another. If a corrector is applied, the responses of the original classifier generate four classes: positive (P), negative (N), false positive (FP), and false negative (FN). As a result, it becomes possible to train two types of correctors: the FP-corrector separating positive and false positive classifier responses, and the FN-corrector separating negative and false negative classifier responses. In the experiments, the VGGish convolutional neural network was used as the initial classifier. The audio signal is converted into a spectrogram and is fed to the input of the neural network which forms the spectrogram feature description and performs a classification. As an example, two ”confused“ classes are selected to demonstrate the increase in classification accuracy. Using the feature description of audio recordings of these classes, an FN-corrector was built, trained and connected to the original classifier. The response from the classifier, as well as the feature description, has been passed to the corrector input. Next, the corrector translated the feature space into a new basis (into a linearly separable space) and classified the classifier answer responding to the question whether the original classifier makes a mistake on such a feature vector or not. If the original classifier made a mistake, then his answer is changed by the corrector to the opposite, otherwise the answer remains the same. The results of the experiments demonstrated a decrease in the level of class confusion and, accordingly, an increase in the accuracy of the original classifier without changing its structure and without collecting an additional data set. The results obtained can be used on IoT devices that have significant limitations on the size of the models used, as well as in solving the problems of domain adaptation which is relevant in audio analytics.

In current digital climate, education sector is evolving as the computer technology advances. Education is being digitized: online classes, online examination methods are conducted, etc. During examination, students are assessed by their answers having given for the question set by a teacher. Today many tools are available to assess the performance of a student using multi choice questions tools which provide instant evaluation, but there are available very limited and operational tools where subjective type answer of students are evaluated. This paper presents a web-based application to address this challenge. It automates the process of subjective answers checking and generates results through using natural language processing methods, like keyword matching semantic, lexical analysis and cosine similarity. Experiments show that appreciated by the teacher result and the system estimation does not have much difference which signifies that the system evaluates answers with a 97 % accuracy. The presented system not only reduces manpower but also eliminates the traditional method of conducting exclusively subjective exams using paper documents. It also eliminates the delays in the paper checking, result generation process. The cases of information leak are being reduced and the objectivity of the assessment is being increased.

The paper presents a methods of image local features extraction research in relation to people authentication problem by face thermogram. As a part of the study, there were formed two datasets for methods training and testing: photographic images and face images in the long-wavelength infrared specter (LWIR) with various factors. The novelty of this study is due to the approach to collecting datasets to verify the accuracy of authentication methods. The dataset was collected with more realistic conditions that affect the quality of authentication, such as changing facial expressions, wearing glasses, medical masks, applying make-up/cosmetics, changing the lighting and temperature conditions of the environment, rotating the head. The methods core is based on the idea of constructing a vector of image features while reducing the dimension and highlighting the boundaries. Likewise, the methods of this group cope well with extracting features task on images and are widely used in the tasks of authentication by 2D face image, as well as in other computer vision tasks. In this paper, four classical methods of local feature extraction are considered: the method of locally binary templates, Gabor wavelets, scale-invariant feature transformation, and Weber’s local descriptor. The classifiers for the feature vectors comparison in this research are SVM and the simplest Perceptron — the basic methods of machine learning. As part of the study, a comparative analysis of each method was carried out in relation to the collected datasets. The methods were trained and tested on a collected face dataset of over 632,000 images of 152 people. As a result of the comparative analysis, it can be concluded that the method of local binary features demonstrates the best result among the considered methods for both types of data: for face thermograms (for SVM — 0,57, for Perceptron — 0,58), for photographic images (for SVM — 0,71, for Perceptron — 0,73). Furthermore, the SIFT method showed similar results: for face thermograms (for SVM — 0,58, for Perceptron — 0,55), for photographic images (for SVM — 0,72, for Perceptron — 0,74). Gabor filters and Weber local descriptor application demonstrate a low accuracy rate in the authentication task by both types of data. The results of the work can be used in access control and management systems to increase the fault tolerance of person authentication. The appliance of the considered methods are effective in the tasks such as processing thermograms for authentication a person by so-called “secondary” signs, for example, by the veins and vessels on face patterns, in cases of facial expressions and appearance changes.

Problems of Wireless Sensor Networks (WSN) are associated with a significant increase in the number of devices on these networks. In this regard, the requirements for the protection and the security of WSN from external influences are increasing significantly. WSN security problems are solved by solving the problem of optimal path routing, energy conservation, and so on. This paper proposes a hybrid model of an efficient packet routing and delivery system to prevent Black-hole attacks. This type of attack is considered the most common on the network due to its unique characteristics. To detect such attacks, a deep learning model using a Convolutional Neural Network (CNN) is proposed. The learning algorithm must be reliable and trustworthy so that attack analysis can be considered at different levels to study the intelligent behavior of network attacks. The paper considers the problem of finding the optimal shortest path using Deep Q-Learning and convolutional neural networks to perform efficient routing and delivery of packets in a safer way. As a result of simulation, the achieved accuracy reached 98.57 %.

The paper proposes a mathematical model of the epidemic process, taking into account the dependence of the rates of cure and loss of immunity on time. Today, mathematical models of epidemics based on the basic Kermack–McKendrick model have become widespread. The most famous models are Susceptible-Infected-Recovered (SIR) and Susceptible-Exposed-Infected-Recovered (SEIR). These models are based on dividing the population into separate groups that are in different epidemic conditions. The description of the models is based on differential equations similar to the equations of birth and death in the process of radioactive transformations of elements in a radioactive chain. However, this approach does not take into account the dependence of the probabilities of the transition of the population from group to group on time spent in the treatment process and in the process of loss of acquired immunity. The known models do analyze the nature of the course of the epidemic for long periods of time, when the process can enter a stationary state. The paper proposes a mathematical model based on dividing the population into separate groups. The first group consists of healthy people susceptible to infection due to contact with members of the second group, which includes the infected population. Members of the third group are being treated, the fourth group includes members of the society who have recovered with antibodies and are vaccinated. The fifth group consists of deceased members of society. In contrast to the SIR and SEIR models, the proposed approach takes into account that immunity is lost over time, and people who survived again move to the group susceptible to infection. The dependences of the probabilities of transition from group to group on time spent both during the treatment process and in the loss of acquired immunity have been taken into account. Thus, the proposed mathematical model is based on five integro-differential equations, two of which are partial differential equations. A new mathematical model has been formulated that makes it possible to take into account the dependence of the cure rate and the probability of transition from the vaccinated state to the initial state on time spent in the corresponding state. It is shown that the proposed model is autocatalytic. With increasing time, a state of bistability is observed, when, under the same boundary conditions, two stable states coexist. Switching between states is determined by the epidemic spread control parameter found in the work. One of the stable states is stationary and leads to the end of the epidemic, the second one leads to the population’s extinction. It has been shown that, for the stationary regime, the form of the distribution function in terms of treatment time and exposure time in the vaccinated state does not affect the final result in any way. The conditions for suppressing the epidemic for managing the process of its development are formulated. Numerical experiments were carried out to simulate the epidemic spreading process, taking into account the constancy of all transition probabilities. Integration of the original system of equations was carried out using the Radaus algorithm for stiff differential equations. The results of numerical simulations have confirmed that the experimental data agree with the theory of the control parameter. The results of the work can be used to organize the management of the epidemic spreading process in order to suppress it as soon as possible by changing the value of the control parameter.

The paper considers the optical systems of virtual reality, which use an optical wedge to eliminate or reduce the conflict of convergence and accommodation (vergence- ccommodation conflict, VAC). The inclined surfaces of the optical wedge introduce specific aberrations into the optical system that can critically affect image quality. The authors analyzed wedge aberrations using relationships derived from the third-order aberration theory. In contrast to existing approaches, the relationships were obtained without the assumption of a small refractive angle or wedge thickness. The results demonstrate that in the scheme with an optical wedge, with characteristics typical for virtual reality systems, the dependence of astigmatism on the angular field is linear and significantly exceeds the other types of aberrations. It is shown that to obtain a high-quality image, the most successful compensation for such aberration can be achieved by using diffractive optical elements (kinoforms) or a Fresnel lens. The paper gives examples of optical systems with a wedge, developed taking into account the requirements for the overall dimensions of virtual reality systems and image quality. The obtained ratios for the aberrations of the optical wedge and the inclined surface make it possible to evaluate the correction capabilities of schemes where such elements are used with arbitrary characteristics. The proposed solutions can be used in virtual reality systems with a reduced value of vergence-accommodation conflict.

The paper considers the problem of controlling the movement of a mobile robot along a given smooth trajectory without measuring its position coordinates. To solve the problem, an adaptive observer of the local coordinates of a moving object is used by measuring the linear speed, yaw angle, and range to a beacon with known coordinates. Then the minimum distance from the robot to the given smooth trajectory is determined. Based on the estimates for the coordinates of the robot and the distance to the curve, we synthesized the control law of the movement along the trajectory with the desired speed under the conditions of uncertainty of the mathematical model. The motion control algorithm is based on the robust sequential compensator method, which ensures that the deviations of the robot from a given trajectory are limited. The proposed coordinate observer ensures asymptotic convergence of the estimation errors to zero. In this paper, we propose two algorithms for determining the minimum distance from the robot to the trajectory: an exact analytical calculation and a nonlinear observer that guarantees the convergence of the estimate to the true value in an arbitrarily short time. The trajectory regulator ensures the movement of the robot along a given trajectory with a limited error. The application of the proposed approach allows one to solve the issues of controlling the movement of a mobile robot without measuring the position coordinates. The approach can be widely applied for controlling self-driving vehicles when they run in tunnels or under a bridge, where it is not possible to measure their coordinates using the satellite navigation systems (GLONASS or GPS).

The development of the medical imaging field is becoming a significant challenge due to the growing need for automated, fast, and efficient diagnostics. Traditionally, blood cells are counted by using a hemocytometer along with other laboratory equipment and chemicals, which is a laborious task. The paper investigates the application of machine learning methods to the identification and classification of blood cells, which allow increasing the recognition rate without deteriorating quality. A comparative analysis of methods for solving the problem of recognizing blood cells based on artificial intelligence approaches is carried out. The paper uses support vector machine, k-nearest neighbors’ algorithm, deep learning (convolutional neural network), and forward propagation neural network. A set of images with cell samples was selected as the initial data for comparison. A comparative analysis of the quality of the considered algorithms was performed on a set of training data with more than 3000 images. It is shown that a program that implements artificial intelligence methods provides a cell recognition time within 4-6 seconds when using an office personal computer, which is significantly less than the time spent by medical workers on one study of a biomaterial. The implementation of the presented results makes it possible to automate the process of studying a biomaterial, reduce the time for conducting and obtaining the result of the analysis of whole blood cells (identification and counting), lessen the influence of operator errors on the result, unload computing resources, thereby increases the efficiency of digital medicine.

Recently, television cameras operating in the near-infrared range have become increasingly widespread. The advantage of shooting in the short wave infrared range is the ability to observe objects in low light and difficult weather conditions. Such cameras can use hybrid sensors that consist of an infrared photocathode and an electron-sensitive matrix based on silicon in a single vacuum volume. The paper investigates the capabilities of one of the latest samples of a similar camera, created at JSC “NRI Electron”. The images of a human, water, ice, snow and other objects in the wavelength range of 0.95–1.7 microns have been analyzed. The images were taken using a television camera based on a hybrid television infrared sensor, which consists of a photocathode and a silicon charge-coupled device in proximity range to each other in a single vacuum volume. Illumination of the objects in the near-infrared range was produced by a continuous spectrum lamp with a maximum of the detected radiation at a wavelength of 1.55 μm. The authors compared the images obtained in the near-infrared range with the ones obtained in the visible region. An explanation is given for the differences between snow images and water and ice images in the near-infrared range. As an example, the difference in light transmission between the surfaces of materials for diving equipment, such as coated and open cell neoprene, is demonstrated. Due to a significant contrast shown in the near-infrared range by images of various objects on the surface of water and ice, it is possible to create an effective system for searching objects on the water. The paper discusses the advantages of the proposed visual search system compared to other systems, including passive systems and the ones operating in the MWIR and LWIR. The research outlined the prospects of using the new camera for building an effective search system for objects and people on water surfaces.

For continuous wavelet transformation, wavelets based on derivatives of the Gauss function are used, and for multiscale analysis, Daubechies wavelets are used. The development of algorithms for forward and inverse continuous wavelet transform in the frequency domain made it possible in this work to synthesize digital filters with a finite impulse response (FIR) different from existing methods. The quality of the synthesized filters was checked by decomposition and subsequent reconstruction of the signals. To do this, several filters were synthesized that completely cover the frequency range of the signal. Since wavelets are bandpass filters, the authors called the filters wavelets. The more precisely the reconstructed signal repeats the shape of the original signal, the better the wavelet constructed by one method or another. A comparison of the accuracy of signal reconstruction shows that the best conversion result is obtained by using synthesized wavelets. The impulse responses of the FIR filters are synthesized so that their frequency response are similar to the frequency responses of wavelets based on derivatives of the Gaussian function of a large order. The greater the filter order, the closer the frequency response is to a square-wave shape. Algorithms for forward and inverse wavelet transformation of a signal in the frequency domain using wavelets based on derivatives of the Gauss function are proposed. Profiling of the program shows that the time of the wavelet transforms using the fast Fourier transform is 15,000 times less than with the direct numerical integration for sampling the signal of 32,768 samples. These algorithms can be used for wavelets with a square-wave frequency response. At the same time, the numerical calculation time is halved. The accuracy of the reconstruction was compared for wavelets based on second-order derivatives, Daubechies wavelets, and wavelets with a square-wave frequency response. The reconstruction accuracy is highest for the latest wavelets. The use of the wavelet construction method is preferable since this method is relatively simple and it is easy to synthesize multiband filters with any form of frequency response. If, when synthesizing using existing methods, a short transition band can be obtained only for long impulse responses, while the transition band using the method of constructing wavelets is absent even for filters of very small orders. The paper presents the impulse responses of twoband, three-band digital filters and their frequency response. FIR digital filters with a square-wave frequency response have a higher delay-band attenuation coefficient compared to existing filters, do not have transition band, and can be used to process one-dimensional and two-dimensional signals.

A method for stabilizing structural anomaly detection under additive noise conditions as well as an algorithm for formal selection of the parameters of the solver rule in the structural anomaly detector based on the Robust Random Cut Forest (RRCF) method are proposed. In the framework of the developed approach, in order to stabilize the process of structural anomaly detection under the influence of additive noise, it is proposed to feed to the input of the RRCF-detector a data stream which is pre-processed by one of the digital filtering methods. In this case, the decision rule for anomaly detection is strictly formalized and transparently interpreted. The selection of parameters of the RRCF-based anomaly detector stabilized by pre-filtering methods of the input data stream is formalized. The RRCF-detector parameters choice within the proposed scheme guarantees a predetermined upper bound for the false alarm probability when deciding to detect a structural anomaly. This property is rigorously proved and formalized as a theorem. The performance of the stabilized RRCF-detector is investigated numerically. The achieved results confirm the performance of the proposed approach provided that the detection threshold is selected in the way proposed in this paper. An example of practical application of the proposed method is presented. The developed approach is promising for the detection of structural anomalies in conditions of observation additive noise, in a situation where it is important to guarantee an upper bound for the probability of false alarm. In particular, the approach can find application in monitoring technological regimes of liquid pumping in pipeline systems or in systems for detecting pre-failure states of technological equipment.

Modern artificial intelligence (AI) technologies are being used in a variety of fields, from science to everyday life. However, the widespread use of AI-based systems has highlighted a problem with their vulnerability to adversarial attacks. These attacks include methods of fooling or misleading an artificial neural network, disrupting its operations, and causing it to make incorrect predictions. This study focuses on protecting image recognition models against adversarial evasion attacks which have been recognized as the most challenging and dangerous. In these attacks, adversaries create adversarial data that contains minor perturbations compared to the original image, and then send it to a trained model in an attempt to change its response to the desired outcome. These distortions can involve adding noise or even changing a few pixels. In this paper, we consider the most relevant methods for generating adversarial data: the Fast Gradient Sign Method (FGSM), the Square Method (SQ), the predicted gradient descent method (PGD), the Basic Iterative Method (BIM), the Carlini-Wagner method (CW) and Jacobian Saliency Map Attack (JSMA). We also study modern techniques for defending against evasion attacks through model modification, such as adversarial training and pre-processing of incoming data, including spatial smoothing, feature squeezing, jpeg compression, minimizing total variance, and defensive distillation. While these methods are effective against certain types of attacks, to date, there is no single method that can be used as a universal defense. Instead, we propose a new method that combines adversarial learning with image pre-processing. We suggest that adversarial training should be performed on adversarial samples generated from common attack methods which can then be effectively defended against. The image preprocessing aims to counter attacks that were not considered during adversarial training. This allows to protect the system from new types of attacks. It is proposed to use jpeg compression and feature squeezing on the pre-processing stage. This reduces the impact of adversarial perturbations and effectively counteracts all types of considered attacks. The evaluation of image recognition model (based on convolutional neural network) performance metrics based was conducted. The experimental data included original images and adversarial images created using attack FGSM, PGD, BIM, SQ, CW, and JSMA methods. At the same time, adversarial training of the model was performed in experiments on data containing only adversarial examples for the FGSM, PGD, and BIM attack methods. Dataset used in experiments was balanced. The average accuracy of image recognition was estimated with crafted adversarial imaged datasets. It was concluded that adversarial training is effective only in countering attacks that were used during model training, while methods of pre-processing incoming data are effective only against more simple attacks. The average recognition accuracy using the developed method was 0.94, significantly higher than those considered methods for countering attacks. It has been shown that the accuracy without using any counteraction methods is approximately 0.19, while with adversarial learning it is 0.79. Spatial smoothing provides an accuracy of 0.58, and feature squeezing results in an accuracy of 0.88. Jpeg compression provides an accuracy of 0.37, total variance minimization — 0.58 and defensive distillation — 0.44. At the same time, image recognition accuracy provided by developed method for FGSM, PGD, BIM, SQ, CW, and JSMA attacks is 0.99, 0.99, 0.98, 0.98, 0.99 and 0.73, respectively. The developed method is a more universal solution for countering all types of attacks and works quite effectively against complex adversarial attacks such as CW and JSMA. The developed method makes it possible to increase accuracy of image recognition model for adversarial images. Unlike adversarial learning, it also increases recognition accuracy on adversarial data generated using attacks not used on training stage. The results are useful for researchers and practitioners in the field of machine learning.

The paper considers the problem of an effective microservices interaction and its organization to support data consistency in fault tolerant and high load systems. The “Saga” microservices orchestration template was used for microservices management. The authors assessed the expediency of asynchronous programming principles usage for designing the Saga coordinator. The simulation of processes in the Saga coordinator was conducted; it considers different specifics of asynchronous and synchronous configurations of distributed transactions (Sagas) management. Synchronous configuration group includes a coordinator with a fixed pool of threads and a coordinator that generates a new thread for each new Saga. Asynchronous configuration group consists of a coroutines based coordinator and a coordinator that uses Linux core scheduler. The set of simulations with different numbers of Sagas and available coordinator processors was executed. It was shown that the use of asynchronous approaches significantly reduces the Saga’s execution duration up to 9.74 times and improves the processor time utilisation value up to 88 %. The obtained data proves the efficiency of asynchronous programming principles applied to the design of the Saga coordinator. The difference in efficiency between the asynchronous algorithms that were implemented in this paper was insignificant. Asynchronous programming principles used to build the Saga coordinator allow it to handle bigger load and to use processor resources in a more efficient way. The outcome of this research can be applied during the design of fault tolerant and high load systems. The paper might be interesting to IT-specialists and researchers focusing on distributed computing.

Methods of studying human motion in computer vision systems can be divided into two types. These are analysis in twodimensional and three-dimensional space. The former uses a single camera image and/ or multiple body sensors. Such an approach leads to a rapid accumulation of error and, consequently, low accuracy of the figure representation. Multiple cameras are usually used in the case of three-dimensional space analysis, while the objects are represented as sets of volumetric elements. Despite the high accuracy of this method, it is associated with high computational complexity and internal network load. The purpose of the paper is to develop a model using a single camera, while approaching threedimensional space analysis methods in terms of accuracy. In this paper a human figure is represented as a skeleton. The skeleton is described by an acyclic connected graph. The general structure of a human figure is analyzed. Fifteen basic points are selected. Physical and logical connections between them were studied and mathematically described. The velocity and spatial characteristics of the points and connections outline the general dynamics of motion. The study describes a model of human motion and gives the option for model construction on the example of a particular image. The developed algorithm for collection and analysis of information estimates relative locations and velocity characteristics of the graph elements. The model can be used for acquisition of information about the reference dynamics of human movements. In case of detecting major differences between the reference and the reality, the behavior is defined as deviant. Thus, the obtained algorithm can be applied in computer vision systems for detection and analysis of human movements.

The paper proposes a method for studying the color rendition of digital cameras. This parameter is conventionally assessed visually by color targets in photography and design. The proposed method compares the chromaticity values of standard test objects with their measured values on a digital camera using special setup features. The setup includes a light source, a collimator for uniform illumination of a reflecting screen with a test object, and a research object, i.e., a digital camera. The camera is positioned at a 45 degree angle to the screen. This setup follows the recommendations of the Commission Internationale de l’Eclairage (CIE) for color measurements. To measure the colorimetric characteristics of the samples, the authors adopted the 0/45 illumination / observation scheme. The test object is illuminated with a light beam, the axis of which makes an angle not exceeding 10° from the normal to the sample surface. The sample is observed at an angle of 45° ± 5° from to the normal. The angle between the axis of the illuminating beam and any of its rays does not exceed 5°. A type A source (color temperature 2856 K) acts as an illuminator. Standardized reference sets of colored optical glasses (light filters) characterized by known chromaticity coordinates were chosen as test objects. To evaluate the results, specialized software has been developed that allows one to select individual pixels, calculate their brightness in order to find the chromaticity coordinates and compare the results with reference values. The method was tested using the Canon EOS 60D digital camera. At the time of measurement, digital filters for correction, anti-aliasing, sharpening, as well as color adjustment were turned off in the camera. The paper presents colorimetric measurements using 58 colored optical glasses. The averaged values fall into seven groups. The selected color space for measurements is sRGB. The measurement results proved the possibility of using the proposed technique for analysing and choosing appropriate digital recording devices for colorimetric measurements in such areas as medicine, chemistry and food industry.

The paper proposes an approach to obtaining images of the objects under investigation based on indirect optical location data. The goal of the study is to increase the graphic similarity of the images and to assign them measuring properties. To achieve this goal, the concept of photogrammetric processing of frame images obtained by conducting indirect optical location in a certain way is formulated. The graphical similarity of the images is proposed to be improved by extracting photometric data related to the object and the background from the registered optical radiation. Based on the selected data, a statistical evaluation of the sample average of the optical radiation intensity from these sources is carried out. The obtained estimates are used to form a monochrome digital image. Adding measurement properties is done by converting the coordinates of the digital image to relative coordinates that have a metric expression. The reason for the decrease in the graphical similarity of the images obtained on the basis of indirect optical location data is determined. In particular, the addition of light waves from different sources, during the allotted exposure time of the photodetector, leads to the merging of the object and the background in the resulting image. The paper presents an approach to the separation of photometric data from different sources that is based on the observation of the phase difference between the emitted and recorded light waves. The authors define the mathematical apparatus for linking the obtained images to the relative coordinate system that is adapted for the case of indirect optical location. The concept of conducting indirect optical location using a special optoelectronic complex is proposed. The study describes the requirements for the equipment of an optoelectronic complex that generates and registers optical radiation with the required parameters. The results of an experiment on the formation of images with measuring properties confirm the feasibility of using the proposed method. Conducting an indirect optical location opens the way to obtaining images of an area that is inaccessible to humans. In particular, the results of the experiment demonstrate that the use of the proposed concept provides images of an object placed behind a light-tight obstacle, which are characterized by the presence of measuring properties and reflect the details of the object under study with high graphical similarity.

The paper presents the results of mining and primary approbation of plasma method for producing pure white corundum. Upgrading ways of pure corundum production is an important task for industry as part of reducing the energy consumption and environmental contamination. The purposes of the research at this stage are as follows: the selection of raw materials, formative evaluation for characteristics of the technology, conducting an experiment on melting and assessment of the sample. The corundum melting is conducted in the reactor using high-voltage plasmatron. Mixture of argon and 25–30 percent of nitrogen is used as the working fluid. The authors suggest using a four-layered protection of a melting reactor in order to ensure both thermal insulation properties and strength characteristics. This is especially relevant under temperature difference of the order of 2000 K and elimination of defective crystallization of the melt from the walls of the reactor. As a result of an experiment on melting alumina marked G-00 using high-voltage air powered plasmatron, the sample with alumina oxide in the amount of 99.79 percent and with absolute hardness equal to 500 was obtained. Further experiments make it possible to determine the prospects of using the proposed technology to obtain samples with an increased content of aluminum oxide. The paper discusses the application of the described technology for industrial production of pure corundum single crystals. The technology will make it possible to obtain samples to be used as abrasives for optical systems and for the production of sapphire glasses and scalpels.

The paper focuses on revealing insider information leaks of financial markets during investment consulting. An original dataset was created, containing the records of the conversations between consultants and clients, presented in the form of dialogs in text format. The applicability of machine learning methods for automating the detection of leaks arising in a conversation between a consultant and a client has been studied. The authors examined the applicability of the following supervised machine learning methods for constructing and training a classifier: probabilistic (Naïve Bayes classifier), metric (k-nearest neighbors algorithm), logical (random forest), linear (support vector machine), and methods based on artificial neural networks. The paper considers various approaches to the construction of a natural language text model, such as tokenization (bag of words, word n-grams: bigrams and trigrams) and vectorization (one-hot encoding). The proposed algorithm for detecting financial markets insider information leaks is based on the use of support vector machine (SVM) and tokenization by bigrams. The obtained results demonstrate that SVM and bigram tokenization provide the highest leakage detection accuracy. The research results can be used in cybersecurity tools development, as well as for the further elaboration of natural language processing methods dealing with information security problems.

The paper considers the problems of modeling and identification of parameters for models of surface ships. The proposed identification method is applied to a modified second order Nomoto model for ship steering. The identification algorithm is based on the Dynamic Regressor Extension and Mixing Method (DREM) that is performed in two steps. At the first stage parameterization is used for a regression model, in which the regressor and regression depend on the measured signals, namely, longitudinal, lateral and angular velocities and steering angle. At the second stage a new regression model is built using linear stable filters and delays. Finally, the parameters are estimated by the standard gradient descent method. The paper proposes a new algorithm which identifies the parameters for models of surface ships. The authors analyzed the prospects of the proposed estimating method by computer experiments. Experiments have shown the advantage of the method: when using the gradient descent method, the transient time spent to estimate the signal parameters is much longer than using the DREM method. At the same time, in the case of using the DREM method, there is no overshoot. The results of the work can serve as a basis for methods, algorithms and software for designing ship automated navigation systems and control systems for other modes of transport. This is confirmed by the simulation results.

The paper presents an analysis of the existing methods for assessing information security risks, their features, advantages and disadvantages, as well as determines the possibility of using such techniques for assessing information security risks in financial institutions. Criteria for comparing information security risk assessment methods have been formed, the advantages and disadvantages of the methods are described. It is shown that, despite the requirements of regulators for assessing information security risks, most of the regulatory documents deal with operational risks. The evaluation of information security risks of credit and financial institutions does not have sufficient regulation and formalization. The authors substantiate the necessity of developing a method for assessing information security risks for credit and financial organizations, taking into account the features of risk assessment inherent to the mentioned organizations. The paper considers the need to create lists of existing threats to the credit and financial sector and their linking to existing vulnerabilities to optimize the process of assessing information security risks. The development of a methodology for assessing information security risks will increase the degree of compliance of credit and financial institutions with the requirements of international, state and industry standards through an optimal set of protection measures and models for evaluating information security risks.

Most of the medical data in hospital information systems databases are stored in an unstructured form. Techniques for processing unstructured records are widely presented in scientific papers focused on English data. This paper proposes a method for intellectual analysis of unstructured allergy anamnesis in Russian in order to identify the presence and type of allergy and intolerance of a patient. The method is based on machine learning algorithms and uses international standards for the exchange of medical data and terminology standards, such as FHIR and SNOMED CT. As a result of the experiment, about 12 thousand medical records were processed. F-measure for the developed classification models ranged from 0.93 to 0.96. The models showed high values of metrics for evaluating the effectiveness of the models. In the future, structured data can be used in models for predicting medical risks. Further development of methods for structuring medical texts will ensure the interoperability of medical data.

The paper examines the issues of increasing the efficiency and reliability of stereo image identification through computational prediction of the position and size of the uncertainty zone in which the desired correspondence point is known to be located. A control point is selected on one of the stereo images, for which it is necessary to find a correspondence point on the second stereo image. Based on the known parameters of the stereoscopic television system and the coordinates of the control point, using the mathematical apparatus proposed in the work, the coordinates of the boundaries of the uncertainty zone on the second stereo image are calculated. The second point of correspondence is found by the search procedure by comparing identical small areas with centers in the control point on the first stereo image and in the points of the uncertainty zone on the second; the comparison is made according to the criterion of minimum quadratic mismatch of intensities. The necessary a priori information for implementing the method is the maximum heights of the relief displayed on stereo images. The ratios of linear dimensions on a flat relief and on an image formed according to the principle of central projection were obtained. Relationships have been obtained that make it possible to obtain, by calculation, the coordinates of the correspondence points and the stereoscopic mismatch for stereo images of a flat relief. For stereo images of a volumetric relief, calculation formulas are obtained for determining the boundaries of the zone of uncertainty in the second stereo image within which the search for the point of correspondence is carried out. The correctness and performance of the obtained relationships are confirmed by computer modeling. Limiting the size of the search area by means of calculated prediction of the uncertainty zone makes it possible to reduce the computational and time costs of the search procedure. Due to this, the efficiency of identifying stereo image points increases and the likelihood of false identification decreases.

The paper presents a study of the effect of laser irradiation of crystalline chalcogenide films of lead selenide (PbSe) on their electrical characteristics caused by irreversible modification of the structure due to valence reconfiguration of lead as a result of its oxidation. The study of the modification features of the electrical properties of the films was carried out because of laser exposure to nanosecond pulses with a wavelength of 1064 nm. Measurements of the electrical characteristics of PbSe films were carried out using the four-probe method. It was shown that when the current was directed parallel to the laser tracks recorded in the darkening mode, the resistance of the modified film decreased by 44 % compared to the original sample, and with the perpendicular direction of the current, the resistance increased by 153 %. The resistance of the film increased more than 27 times after laser irradiation in the bleaching mode, regardless of the direction of the current relative to the laser tracks. The experimentally measured temperature and its gradient along the laser spot on the film in the darkening and bleaching modes turned out to be in good agreement with the proposed mathematical model of the thermal effect of laser pulses. It has been shown that the processes of laser modification of films occur at lower temperatures than during standard heat treatment in a furnace. The obtained results can be applied in the development of photodetectors in the middle IR range of the spectrum based on PbSe film.

Coriolis vibration gyroscopes are a class of promising inertial primary information sensors that respond to the rotation of the resonator base through Coriolis inertial forces arising in the vibrating shell. Currently, two directions for the production of resonators for such gyroscopes have been developed: from quartz glass, a material with extremely low internal friction, and based on the processing of a metal alloy. When using the first direction, thanks to the high quality factor of quartz, it is possible to create navigation-class integrating gyroscopes. Existing samples of Coriolis vibration gyroscopes with metal resonators, as a rule, are angular velocity sensors. The problem of creating an integrating mode of a gyroscope with a metal resonator is associated with the low quality factor of metal alloys which usually does not exceed 35,000. With this value of quality factor, the duration of operation of the gyroscope in the angular deviation sensor mode will be several seconds. The paper presents methods for ensuring the functioning of Coriolis vibration gyroscopes, including the integrating gyroscope mode. A mathematical description of Coriolis vibration gyroscopes with a cylindrical cavity resonator is given based on the dynamic model of Dr. D. Lynch using the method of envelope amplitudes of oscillations. The mathematical model is supplemented with corrections that provide compensation for the dissipation of the resonator oscillations energy to implement the integrating mode of the gyroscope. The conditions for complete compensation of vibration energy dissipation are shown. A description of methods for exciting a standing wave in a resonator using periodic forcing and by creating self-oscillations is presented. It is shown that the duration of the transient excitation process is determined by the time constant of the resonator. The results of experimental studies of Coriolis vibration gyroscopes with a low-Q metal resonator are presented confirming the possibility of implementing an integrating mode of operation of the gyroscope. The initial excitation of the resonator oscillations is carried out by a self-oscillating circuit. According to the results of experimental studies, the quality factor of the metal resonator was increased by a factor of 17. The operating time of Coriolis vibrating gyroscopes has been equally increased. The possibility of constructing Coriolis vibration gyroscopes in the integrating gyroscope mode based on a low-Q metal resonator has been shown theoretically and experimentally. The solution to this problem was based on a circuitry method for increasing the quality factor. In principle, the quality factor of the resonator can be significantly increased compared to the figure achieved in the experiment. This will ensure a longer operating time of Coriolis vibration gyroscopes in the integrating mode.

It is widely known that among sound signals generated by natural and anthropogenic phenomena, the most long-lived are waves of frequency less than 20 Hz, called infrasound. This property allows tracking at a distance by infrasound monitoring the occurrence of high-energy events on regional scales (up to 200–300 km). At the same time, the separation of useful infrasound signals from background noise is a non-trivial task in real-time and post-facto signal processing. In this paper we propose a new method for classification of specific signals in infrasound monitoring data using Shannon permutation entropy and vectors of frequency distribution of occurrence frequencies of permutations of consecutive sample values of rank 3 (number of permutation elements). To evaluate the validity of the proposed entropy-based classification method, two machine learning methods — random forest method and classical neural network approach — implemented in Python language using Scikit-lean, TensorFlow and Keras libraries were used. The classification quality was evaluated against the traditional frequency-based method of class extraction based on Fourier transform. Recognition was performed on the prepared infrasound monitoring data in the Altai Republic. The results of computational experiment on the separation of 5 classes of signals showed that classification by the proposed method gives the same results of recognition by neural network with in comparison with frequency classification of the original data; the recognition accuracy was 51–58 %. For the random forests method, the recognition accuracy of frequency classes was slightly higher: 51 % vs. 45 % for classes using the permutation entropy method. The analysis of the results of the computational experiment shows sufficient competitiveness of the method of classification by permutation entropy in the recognition of infrasound signals. In addition, the proposed method is much easier to implement for inline signal processing in lowconsumption microcontroller systems. The next step is to test the method at infrasound signal registration points and as part of the infrasound monitoring data processing system for real-time event detection.

The most important method of ensuring data integrity is correcting errors that occur during information storage, processing or transmission. The error-correcting coding methods are used to correct errors. In real systems, noise processes are correlated. However, traditional coding and decoding methods use decorrelation, and it is known that this procedure reduces the maximum achievable characteristics of coding. Thus, constructing computationally efficient decoding methods that would correct grouped errors for a wide class of codes is an actual problem. In this paper the decoding by information sets is used to correct single bursts. This method has exponential complexity when correcting independent errors. The proposed approach uses a number of information sets linearly growing with code length, which provides polynomial decoding complexity. A further reduction of the number of information sets is possible with the proposed method of using dense information sets. It allows evaluating both the set of errors potentially corrected by the code and the characteristics of the decoder. An improvement of the decoding method using an error vector counter is proposed, which allows in some cases to increase the number of corrected error vectors. This method allows significantly reducing the number of information sets or increasing the number of corrected error vectors according to the minimum burst length criterion. The proposed decoders allow correction of single error bursts in polynomial time for arbitrary linear codes. The results of experiments based on standard array show that decoders not only correct all errors within the burst correcting capability of the code, but also a significant number of error vectors beyond of it. Possible directions of further research are the analysis of the proposed decoding algorithms for long codes where the method of analysis based on the standard array is not applicable; the development and analysis of decoding methods for multiple bursts and the joint correction of grouped and random errors.

The results of a study of the luminescent dynamics of combined aqueous-alcoholic solutions of extracts of red viburnum (Viburnum opulus L.) fruits with chitosan and gold nanoparticles at different oxygen concentrations were presented. The search for natural sources of photosensitizers is an urgent task. The sensitive analytical methods, in particular luminescence, with amplification of the analytical signal as a result of the generation of plasmons in nanoparticles of noble metals, have been used in this work. An additional studies under the conditions of plasmon energy generation showed a significant changes in the dynamics of optical spectra with variations in oxygen concentration in solutions. The spectral-temporal dynamics was investigated with complete oxidation of vitamin C in the studied system. The main method for recording the dynamics of interaction of Viburnum opulus L. flavonoids with oxygen molecules is the luminescent method. Luminescence spectra were measured by means of Fluorolog-3 optical system (Horiba, Japan). Methods of absorption analysis (Shimadzu spectrophotometer, Japan) were also used in the work. The nanosecond luminescence lifetimes of the extracts were measured in the multichannel photon counting mode using a picosecond NanoLED-405L nanoled by means of Fluorolog-3 spectral setup. Microsecond lifetimes were recorded when excited by a pulsed Xe lamp. For the synthesis of gold nanoparticles, the method of laser ablation of a metal plate of gold in distilled water was used. Laser ablation was performed at the LQ929 installation of Solar Laser System (Belarus). A plasmonic effect of amplification of the optical absorption density and luminescence intensity was detected. The kinetics of luminescence quenching of Viburnum opulus L. extract with chitosan under the influence of gold nanoparticles, close to diffusion, was studied. The oxygen concentration at which the flavonoids of the extract are oxidized was spectrally determined. Under the oxygen concentration changing, the dependences of changes in the luminescence intensity of the extract with chitosan at the wavelengths of registration of luminescence spectra were established. When oxygen was doped into all solutions, the spectral and kinetic features of luminescence attenuation with maxima at wavelengths of 480 and 580 nm were detected and investigated. It was established that the lifetime of luminescence at a registration wavelength of λ = 480 nm varies depending on the concentration of gold nanoparticles and the concentration of oxygen molecules and it is the nanosecond spectral region (3–4 ns). It was shown that luminescence at a wavelength of 580 nm is due to the oxidized form of quercetin which is a part of the Viburnum opulus L. flavonoids, appeared at a high oxygen concentrations. Long-lived chemiluminescence at a wavelength of 580 nm with time decay of 15 μs as a result of radical processes involving molecular oxygen and extract molecules was recorded. The spectral methods presented in this paper, as well as a method for determining of quercetin as a result of oxygen oxidation of flavonoids of red viburnum fruits, can be used in the field of biophysics, biotechnology and chemical analysis.

We report the ZnO/ZnS and ZnO/ZnSe nanocomposites synthesized using the solvothermal-microwave method. Raman analysis was thoroughly studied to explain phonon vibration mode in this paper. The strong intensity confirms the highfrequency phonon mode of hexagonal wurtzite ZnO. Also, the presence of Raman intensity of the cubic ZnS and ZnSe structures indicates the longitudinal optical phonon mode. In addition, we find several slight shifts in all ZnO modes for ZnO/ZnS and ZnO/ZnSe which demonstrate stress and strain in the crystal lattice. We investigate the change in particle size from confocal Raman microscopy. Therefore, the modifications to the material structure and particle size have enhanced its characteristics. Accordingly, the nanocomposite heterostructures by the simple chemical method are attractive materials suitable for optoelectronic devices.

This paper presents the design of a microstrip patch antenna based on a Square Split Ring Resonator (SSRR). Wireless technology is switching from 4G to 5G due to the need to overcome limitations, such as low throughput, high latency and path loss. To increase data transfer speeds, the next generation of wireless networks uses 5G terahertz technology. The use of microstrip patch antennas in wireless technologies has increased significantly due to their low cost and simplicity of design as well as the ease of printed circuit board fabrication. However, in some cases their use is limited by low bandwidth, low gain and low throughput. To solve these problems, the Fennec Fox optimization algorithm is used. The algorithm allows you to optimize the length of the microstrip patch antenna resulting in increased gain and reduced return loss. Bakelite is used as a substrate. The width of the patch antenna is set according to the most suitable length selected. To increase the bandwidth and Voltage Standing Wave Ratio (VSWR), a square split ring resonator (SSRR) is used as a metamaterial. An evaluation of the designed microstrip patch antenna model with existing patch antennas was performed. The estimated values of the parameters of the proposed model were the following values: return loss –72.54 dB, resonant frequency 1.11 THz, achieved gain 15.25 dB, VSWR value 1.5646. The estimated values of the developed model exceed those of existing samples. Thus, the developed microstrip patch antenna using Fennec Fox optimization and square split ring resonator metamaterial shows better results in the terahertz range.

Despite the widespread use of automatic speech recognition and video subtitles, sign language is still a significant communication channel for people with hearing impairments. An important task in the process of automatic recognition of sign language is the segmentation of video into fragments corresponding to individual words. In contrast to the known methods of segmentation of sign language words, the paper proposes an approach that does not require the use of sensors (accelerometers). To segment the video into words in this study, an assessment of the dynamics of the image is used, and the boundary between words is determined using a threshold value. Since in addition to the speaker, there may be other moving objects in the frame that create noise, the dynamics in the work is estimated by the average change from frame to frame of the Euclidean distance between the coordinate characteristics of the hand, forearm, eyes and mouth. The calculation of the coordinate characteristics of the hands and head is carried out using the MediaPipe library. The developed algorithm was tested for the Vietnamese sign language on an open set of 4364 videos collected at the Vietnamese Sign Language Training Center, and demonstrated accuracy comparable to manual segmentation of video by an operator and low resource consumption, which will allow using the algorithm for automatic gesture recognition in real time. The experiments have shown that the task of segmentation of sign language, unlike the known methods, can be effectively solved without the use of sensors. Like other methods of gesture segmentation, the proposed algorithm does not work satisfactorily at a high speed of sign language when words overlap each other. This problem is the subject of further research.

Scheduling of computing operations takes an important place in the process of distributed information processing and control systems design, especially in conditions of limited energy resources of the system. This becomes especially important for computers located on autonomous carriers, such as unmanned aerial vehicles, autonomous underwater vehicles, etc. The energy resources in such systems are limited that leads to high requirements for the energy efficiency of the carrier systems including computing ones. The paper presents the job scheduling method for a distributed computing system on a chip which allows reducing the power consumed by the system. The proposed task scheduling method includes two stages. At the first stage, jobs are assigned with the determination of an energy-efficient architecture of the system characterized by the minimum power consumption. At the second stage, jobs are scheduled taking into account the criterion that minimalizes the average job implementation time. A feature of the problem being solved in this case is the necessity of job scheduling in the system with more than one information output which does not allow applying any of the known scheduling methods to the system. The first stage of the proposed method is implemented by implementation additional processors with a simultaneous decrease in the clock frequency and supply voltage. For the second stage of the method, the job scheduling algorithm is proposed which involves the preliminary construction of a private schedule for each output of the system with further integration of these schedules into the general schedule using a heuristic procedure. The scheduling algorithm functioning is illustrated by an example of a solution for a simple system. The advantage of the proposed heuristic method is the possibility of scheduling calculations, taking into account criteria of the minimum power consumption and the minimum average residence time of a task in the system simultaniously. This makes it possible to increase the energy efficiency of solving problems in distributed computing systems on a chip, which contributes to increasing the autonomy of systems in which they are used in. The proposed algorithm has polynomial complexity, therefore, due to the relative simplicity of the algorithm, it can be used for scheduling and rescheduling jobs in real time for complex systems.

In this paper, we have developed a system for assigning customers to the routes based on their emotional state and age in Public Service Systems (PSSs). The Squeeze-and-Excitation (SE) method was used to develop the models, it improves the efficiency of the Deep Convolutional Neural Networks (DCNN) architecture by increasing the information flow between layers and enhancing important features. The method is based on compressing and exciting information at each convolution stage, which allows obtaining a vector of channel importance scores and using it to reweight the channels of the feature map. The study showed that this method allowed improving the quality of classification and reducing the model training time. The model of emotional target routing was developed based on the Newton interpolation polynomial to route customers based on their emotional state and age. The interpolation function in this model calculates the waiting time for customers according to their emotional state. Three models of binary classification of emotions and ages were developed, namely, two models for recognizing the emotional state of the customer, and one model for recognizing their age. The first and third models utilize DCNN from scratch using the new SE approach based on the attention mechanism. The second model uses the Support Vector Machine (SVM) method. The evaluate method was used to test the model after training, which allows evaluating the quality of the model on new data that was not used during training. This is done to check how accurately the model can predict the values of the target variable on new data. The evaluate method utilizes quality evaluation metrics such as accuracy, recall, and F1-score to assess the performance of the model. According to the experimental data obtained, the first and the second developed models achieved the validation accuracy of 72 % and 66 %, respectively, on the FER-2013 and Adience datasets. Their sizes were 0.69 MB and 369 MB, respectively. At the same time, the age recognition model achieved the accuracy of 88 % with the size of 1.68 MB. The mathematical model of emotional target routing (TERSS) was developed to minimize conflicts in public service systems. The developed system can automatically route customers based on their emotional state (presence of anger) and age to the appropriate operator. Thus, customers over 60 years old or with the anger level of 60–80 % are directed to a senior operator who knows how to communicate with elderly or emotionally excited customers, while customers with the anger level of 80–100 % are directed to a psychologist. This research can be applied in PSSs to detect the features of customers’ age and anger. Moreover, it can be applied in various areas where there is a contact with a large number of people, such as banks, supermarkets, airports access control systems, police stations, subways, and call centers.

To operate in unstructured environments, robots must have the property of passivity which can be realized either through control algorithms or physical elastic elements. Flexible elements can be used to recover energy, absorb peak shock loads, and simplify the control system, generally reducing the requirements for accurate information about the robot’s environment. The modeling of flexible bodies, for example using the finite element method, is computationally demanding, which limits the simulation of the dynamic behavior of robots with flexible elements. In this paper, we propose an approach for analytical and simulation modeling of flexible joints using the planar case of a spatial spring model, which provides high speed simulation without loss of accuracy. The synthesis of the flexible joint model consists of numerical optimization of the nonlinear stiffness diagrams of the rotational and translational degrees of freedom for the planar case of the spatial spring model. The synthesized flexible joint model allows describing the relative motion of two links. In the first step of the synthesis, the flexible joint is optimized by finite element method to find the reference data of applied load and corresponding deformations. In the second stage, an optimization problem is solved to find nonlinear stiffness diagrams for the planar case of the spatial spring model; the criterion is to minimize the error between the reference data and the optimized spring model. In the third stage, the obtained results are verified by simulation and/or physical experiment. The method of analytical and simulation modeling of flexible joints with the help of spatial spring model is proposed, the procedure of optimization of stiffness of spring model is proposed, verification in simulation environment is carried out, full-scale experiment is carried out, comparison of simulations by finite element method, simulation with the help of spring model and results of full-scale experiment is provided. The proposed method allows the calculation of a simulation model of a flexible joint approximately twice as fast as the finite element method. The proposed model of flexible joint is necessary to increase the speed of simulation modeling of mechatronic and robotic systems with compliant hinges without loss of accuracy. Approbation of the method is planned for the design of locomotion, manipulation, wearable robots, and gripper devices.

In the search for new magnetically ordered phases of materials, solid-state synthesis technologies in reaction crucibles are used. The final result of the synthesis process in reaction crucibles is conditioned, in particular, by technological factors, the mode of current flow and its density, the achieved temperature in the reaction zone, exposure time, geometrical parameters of the crucible and the reaction zone, etc. The paper presents the results of influence investigation of the reaction volume filling degree with tin melt on the processes of heat and mass transfer during its electrothermal treatment. A model describing diffusion processes in the reaction zone during the synthesis of iron and tin intermetallides under electrothermal treatment has been proposed. The diffusion process in the reaction crucibles of the iron-tin system was investigated by the finite element method in the Comsol Multiphysics software environment. It is shown that the decrease in the degree of filling of the reaction crucible with synthesis components leads to a change in the distribution of current density and a decrease in the temperature in the reaction zone, which affects the mass transfer processes. The results of the work can be used in the analysis of experimental data on the production of intermetallides by reaction synthesis and determination of the necessary technological parameters for the synthesis of new materials.

In the paper, the problem of compensation of external disturbance in multichannel systems with unmeasurable state vector and delay in the control channel is considered. It is assumed that the disturbance has a harmonic form. To solve the problem of estimating the state vector of a system, a full-order observer with Unknown Input Observer is constructed. A new observer of external disturbance is formed on the basis of the state vector estimates produced by this observer. As a result, a system is formed that uses estimates with an extended state vector. For this system, a regulator is constructed that provides compensation of the disturbance. The proposed algorithm guarantees boundedness of all signals in the closed-loop system and asymptotic stability of the output. It does not require identification of parameters of external disturbance. To demonstrate the performance and efficiency of the proposed approach, computer simulation using MATLAB Simulink software environment is performed. The developed algorithm can be effectively applied in systems with external disturbance in the form of sinusoidal signals, including systems exposed to wind, ship systems, motion control systems of robotic complexes of various types, and others.

The problem of optimizing the choice of parameters for installing a video camera, such as the location and viewing angles, tilt and pan to increase the information content of the generated video signal, is considered. The relevance of the paper is due to the lack of methods and programs for automating the process of choosing these parameters. The problem is solved when the pixel density is reached, which is necessary for solving the task of observation. It is based on the proposed model for representing view areas, surveillance and camera locations as discrete sets in accordance with the observation task being solved, which determines the required minimum pixel density as well as selected criteria and restrictions. It gives the opportunity to solve the problem programmatically, unlike existing solutions that use empirical approaches. The main and additional criteria as well as limitations are formulated according to which it is possible to optimize the position of the camera relative to the required surveillance area — the observation task to be solved, the minimum required camera resolution and the maximum information content of the generated image. Algorithms for calculating estimates of the near, far and side boundaries of the view area as well as view angles, pan and tilt are formulated. The adequacy of the proposed model to real areas of observation, review and location of cameras is substantiated. An example of solving an optimization problem is given, which confirms the correctness of using the proposed method. The results obtained make it possible to automate the design process and minimize the influence of the human factor when choosing the location and installation parameters of cameras in the process of designing surveillance systems. The results of the work can be used in the development of algorithms and programs for computer-aided design of surveillance systems.

Generating a realistic three-dimensional model of the human body is a very time-consuming task. Even with the necessary computing resources, generation errors occur on the figures of people who differ from the average physique. In this paper, an experimental algorithm for reading anthropometric data from only two full-face and profile photographs is proposed. The proposed solution to the problem of generation using the selection of anthropometric points involves setting the constraints of the SMPL (Skinned Multi-Person Linear Model) model. For segmentation of the human body based on empirical studies, a modification of the Fully Connected Convolutional Neural Network (FCN) ResNet101, trained on the COCO Segmentation 2017 dataset, was used. With its help, the basis for the detection of anthropometric points in full-face and profile photos was obtained. The error in determining anthropometric points ranges from 2 to 5 % depending on their location. The constraints for the SMPL rendering model are calculated using the Levenberg- Marquardt algorithm. For its correct operation, a special cost function is proposed, taking into account the features of this task. The dataset collected by the authors of the article (117 people of different physiques and height) shows that the proposed method allows you to obtain a small average absolute error (MAE = 0.0395 m) and a high coefficient of determination (R2 = 0.913). The graph of anthropometric points sets stricter conditions for generating a figure and any deviation from the graph is a consequence of a large generation error. The proposed solution allows you to accurately generate a model of the human body. At the same time, low requirements for computing resources and the quality of users’ initial photos remain. The proposed solution can be used in online fitting rooms, which adds additional complexity to the task due to the requirements to restore the figure from only two pictures as well as the need to accurately reproduce the features of male and female figures.