The article discusses the topic of high-availability multiprocessor systems used in tasks such asgeolocation, targeting, atomic systems, forecasting, surveillance, tracking and others. When emergencysituations arise, such as a malfunction or failure of individual processor modules of the system, as wellas situations associated with operational impact on a multiprocessor system, there is a need for an urgentresponse. A multiprocessor system can respond to emergency situations in a certain way, whichconsists of scheduling the placement or relocation of parallel tasks. The placement planning problem isformally defined as the process of mapping the vertices and arcs of a weighted digraph describing thetasks being performed onto an irregular graph, which in turn represents the physical structure of a multiprocessorsystem. When choosing the optimal transformation, special attention is paid to minimizingthe total weight of the arcs that reflect the relationships between completed tasks. This process is essentiallya more complex version of the graph search problem. It is important to emphasize that this type ofsearch is a classical NP-complete problem in graph theory. Beehive algorithms, genetic evolution, antcolonies, and guillotine cutting are all popular methods for finding optimal placements that are notsuitable for this task because they mostly perform the search at the software level. In order for the systemto quickly respond to emergency situations, it must quickly perform calculations, which these methodscannot allow. Therefore, an urgent task is to develop a method and algorithm for planning theplacement of tasks in matrix hypercubic multiprocessor systems of high availability. This work continuesthe ideas presented in previously published works in this area in terms of combining search and calculationsteps to test intermediate options. Additional information in the form of relations of distances betweengraph elements allows us to reduce the search, which is confirmed by testing on typical graphs.

Purpose of research. Computing networks of a plurality of processor modules have existed for a long time, however, with the development of wireless technologies, the possibility of high-quality and inexpensive implementation of the idea of real-time reconfigurable computing networks (RRTCN) has emerged. The purpose of this work is to select a wireless data transmission network for the RRTCN and create an algorithm for the dynamic selection of the "master" processor module.Methods. The research methods of the work are based on the definitions of the theory of sets and graphs. In particular, a grapho-theoretical approach to distribution is taken as a basis, supplemented by the introduction of a system of criteria for a reconfigurable real-time computing system based on a wireless data transmission protocol.Results. In this paper, the most suitable data transfer protocol for the local RCSRV is selected, a developed mathematical model of changing a real-time computing system built on a wireless protocol is presented, and an algorithm for selecting the processor module (PM) of the RCSRV "master" is developed. The novelty of the work is the use of the distance matrix for the mathematical description of the change in the RRTCN and its use to determine the PM of the "master" in the aircraft.Conclusion. The developed algorithm for choosing the PM of the "master" RRTCN and the mathematical model of changing the position of the PM in a real-time computer system built on a wireless protocol allows organizing the optimal choice of the PM of the "master" in a reconfigurable computing system. On the basis of the presented mathematical model, in the future, it is possible to create a full-fledged mathematical model of the RRTCN and an algorithm for placing tasks in it.

Рассмотрены вопросы синтеза систем логического управления в базисе логических мультиконтроллеров. Приведен ряд дискретных комбинаторных оптимизационных задач, возникающих в процессе синтеза, наряду с методами и алгоритмами их решения в рамках синтеза разбиений граф-схем параллельных алгоритмов логического управления, оценки их асимптотических сложностей, результаты сравнительного анализа и их влияние на аппаратную сложность синтезируемых систем. Предназначена для научных работников и инженеров, занимающихся вопросами проектирования систем логического управления, а также аспирантов, магистрантов и студентов старших курсов, обучающихся по направлениям подготовки 09.03.01 Информатика и вычислительная техника, 09.04.01 Информатика и вычислительная техника.

The paper substantiates the use of multispectral optoelectronic sensors intended to solve the problem of improving the positioning accuracy of autonomous mobile platforms. A mathematical model of the developed device operation has been suggested in the paper. Its distinctive feature is the cooperative processing of signals obtained from sensors operating in ultraviolet, visible, and infrared ranges and lidar. It reduces the computational complexity of detecting dynamic and stationary objects within the field of view of the device by processing data on the diffuse reflectivity of materials. The paper presents the functional organization of a multispectral optoelectronic device that makes it possible to detect and classify working scene objects with less time spending as compared to analogs. In the course of experimental research, the validity of the mathematical model was evaluated and there were obtained empirical data by means of the proposed hardware and software test stand. The accuracy evaluation of the detected object, at a distance of up to 100m inclusive, is within 0.95. At a distance of more than 100 m, it decreases. This is due to the operating range of a lidar. Error in determining spatial coordinates is of exponential character and it also increases sharply at a distance close to 100 m.

The paper discusses a method of visual attention based on vertex ranking of graphs on the basis of image features. The aim of the research is to develop a method that allows high-precision detection of objects in images with low color contrast between the selected and background areas. The image is pre-segmented into regions to calculate the saliency map. The graph is based on regions. Each region is associated with related regions, as well as with areas adjacent to adjacent regions. The regions are vertices of the graph. The vertices of the graph are ranked according to the characteristics of the corresponding image areas. The scope is highlighted based on requests from background areas. The saliency map is determined based on background area queries. Regions adjacent to the edges of the image belong to the background areas. Color features of the image were used in the existing approach of visual attention based on the manifold ranking. Texture features and shape features are additionally used in the proposed method to improve accuracy. Gabor's energy function is used to calculate texture features. The distance between centers of the regions is calculated by analyzing the form. The proposed method has shown good results for detecting objects in images in which the background color and object color are in similar ranges. The experimental results are presented on test images. Precision-recall curves showing the advantage of the developed method are constructed.

Рассмотрена задача составления расписания загрузки процессоров в мультипроцессорных системах критического назначения.Предложен метод планирования загрузки, ориентированный на аппаратную реализацию в связи с невозможностью применения программных средств.Описывается функциональная схема специализированного устройства планирования загрузки, проведен анализ его аппаратной и временной сложности.Путем суммирования времени выполнения операций элементами устройства с учетом увеличения возможного числа процессоров в системе получена диаграмма зависимости, из анализа которой можно сделать вывод, что время работы устройства планирования за

A method of feature matching in images using descriptor structures is considered in the work. The descriptors in the developed method can be any known solutions in the field of computer vision. However, inaccuracies can occur when matching image pairs. It is proposed that descriptor structures should be compared to eliminate the “outliers”. Descriptor structures are described using graphs. An Umeyama method is used to find matching features using descriptor structures. The method is based on the decomposition of matrices into eigenvalues and eigenvectors for weighted graph matching problems. Thus, matches are based on the descriptor at the initial stage. Two graphs are then constructed for each image based on the resulting sets of mapped features. The weights of the graph are distances between all image features, calculated using the Gauss function. Weight matrices are built for each graph. Matrix decomposition is carried out into eigenvalues and eigenvectors. The resulting matrix is calculated based on the Umeyama method and correct matches are found. Thus, false matches are excluded from the set of matches obtained using descriptors by comparing structures. The method is invariant to zoom and in-plane image rotation. The method leads to correct results only if the number of correct matches is greater than the number of false matches. The complexity of the developed algorithm is proportional to the number of matches found.

Purpose of research . Th е t а sk о f со m р iling а р l а n f о r l оа ding р r осе ss о rs in multi р r осе ss о r syst е ms о f с iti са l р ur ро s е (tr ас king, о bs е rv а ti о n, а iming, а t о mi с а nd fin а n с i а l syst е ms) is dis с uss е d. M о d е rn а lg о rithms im р l е m е nt th е р r о bl е m m а inly р r о gr а mm а ti са lly а nd а r е un ассер t а bl е f о r th е со nsid е r е d с l а ss о f syst е ms b еса us е о f th е l о ng s о luti о n tim е , whi с h l еа ds t о а d ес r еа s е in its а v а il а bility а nd а со rr е s ро nding d ес r еа s е in ре rf о rm а n се . Existing арр r ос h е s, а s а rul е , sh о rt-t е rm р l а nning а r е о ft е n f ос us е d о n th е r ар id r е s ро ns е о f о ng о ing р r осе ss е s, whi с h а ls о d ое s n о t а ll а w t о ас hi е v е th е r е quir е d ре rf о rm а n се . А s а result, th е а rti с l е р r оро s е s а n а lg о rithm f о r s с h е duling th е l оа ding о f р r осе ss о rs in multi р r осе ss о r syst е ms, f ос us е d о n h а rdw а r е im р l е m е nt а ti о n, using th е а dditi о n а lly intr о du се d m а tri се s о f tim е , о rd е r а nd о rd е r. Methods . Th е а rti с l е р r оро s е s е bl ос k di а gr а m а nd а fun с ti о n а l di а gr а m о f а s рес i а liz е d d е vi су f о r р l а nning th е l оа ding о f р r осе ss о rs о f о multi р r осе ss о r syst е m. А distin с tiv е f еа tur е is th е us е о f th е d е v е l оро d d о vi се а s а n а dditi о n а l ассе l е r е t о r со nn ес t е d t о th е ра r а ll е l ро rt о f th е со m р ut е r. Th е р r оро s о d d е vi се in this са s е ре rf о rms а ll th е са l с ul а t е d fun с ti о ns о f th е р r оро s е d а lg а rithm, th е r е by r е du с ing th е m а in l оа d о n th е со m р uting syst е m, th е r е by r е du с ing th е t о t а l l оа d. А stru с tur а l а nd fun с ti о n а l di а gr а m о f th е d е vi се f о r s с h е duling а l оа ding s с h е dul е is giv е n, whi с h а ll о ws th е со nstru с ti о n о f со rr е s ро nding tim е di а gr а ms о f t е m ро r а ry а nd h а rdw а r е со m р l е xity. R е sults . Fr о m th е а n а lysis о f th е gr ар hs о f th е d ере nd е n се о f th е in с r еа s е in th е h а rdw а r е со m р l е xity о f th е d е vi се о n th е numb е r о f р r осе ss о rs, а n е x ро n е nti а l in с r еа s е in th е оре r а ting tim е о f th е d е vi се , d уру nding о n th е numb е r о f р r осу ss о rs, f о ll о ws. Du е t о th е е x ро n е nti а l gr о wth in th е numb е r о f р r осе ss о r l оа ding ор ti о ns, th е h а rdw а r е im р l е m е nt а ti о n о f th е s с h е duling а lg о rithm sh о ws а n а dv а nt а g е . Con с lusi о n . With th у h у l р о f th е р r оро s у d а lg о rithm а nd d у vi се f о r р l а nning р r осе ss о r l оа ding, it is ро ssibl е t о in с r еа s е th е s рее d о f l оа d со m р il а ti о n, d ес r еа s е th е t о t а l а m о unt о f со mmuni са ti о n d о l аа y а nd in с r еа s е th е t о t а l ре rf о rm а n се о f multi р r осе ss о r syst е ms.

A problem of automatic comparison of spatial objects on maps with different scales for the same locality is considered in the article. It is proposed that this problem should be solved using methods of topological data analysis. The initial data of the algorithm are spatial objects that can be obtained from maps with different scales and subjected to deformations and distortions. Persistent homology allows us to identify the general structure of such objects in the form of topological features. The main topological features in the study are the connectivity components and holes in objects. The paper gives a mathematical description of the persistent homology method for representing spatial objects. A definition of a barcode for spatial data, which contains a description of the object in the form of topological features is given. An algorithm for comparing feature barcodes was developed. It allows us to find the general structure of objects. The algorithm is based on the analysis of data from the barcode. An index of objects similarity in terms of topological features is introduced. Results of the research of the algorithm for comparing maps of natural and municipal objects with different scales, generalization and deformation are shown. The experiments confirm the high quality of the proposed algorithm. The percentage of similarity in the comparison of natural objects, while taking into account the scale and deformation, is in the range from 85 to 92, and for municipal objects, after stretching and distortion of their parts, was from 74 to 87. Advantages of the proposed approach over analogues for the comparison of objects with significant deformation at different scales and after distortion are demonstrated.

The article considers the fuzzy model for navigation of a mobile robot operating in two modes. In the first mode the mobile robot moves along a line. In the second mode, the mobile robot looks for an target in unknown space. Structural and schematic circuit of four-wheels mobile robot are presented in the article. The article describes the movement of a mobile robot based on two modular neuro-fuzzy system. The algorithm of neuro-fuzzy inference used in two modular control system for movement of a mobile robot is given in the article. The experimental model of the mobile robot and the simulation of the neuro-fuzzy algorithm used for its control are presented in the article.

We discuss the development of a structural descriptor for object detection in images. The descriptor is based on a graph, whose vertices are the centers of mass of segment features. The embedding of the graph in a vector space is implemented using a Young-Householder decomposition and based on differential geometry. Compound curves are used to describe the relationship between the points. The image graph is described by a matrix of curvature parameters. The distance matrix for the graphs of the candidate object and the reference object is calculated using the Hausdorff metric. A multidimensional scaling method is used to represent the results. Images of test objects and images of human faces are used to study the developed approach. A comparison of the developed descriptor with the Viola-Jones method is performed when detecting a human head in the image. The advantage of the developed approach is the image rotational invariance in the plane while searching for objects. The descriptor can detect objects rotated in space by angles of up to 50 degrees. Using the mass centers of segments of features as the graph vertices makes the approach more robust to changes in image acquisition angles in comparison with the approach that uses image features as the graph vertices.

The article describes the peculiarities of using weighted random selection method when solving a task of searching for sub-optimal partitions of parallel algorithms flowgraphs-that arises in designing systems of logical control in the basis of logical multicontrollers (LMC). A review of known methods of solving the problem is presented. Most of these methods are ‘greedy’ successive methods (except random selection method) and provide solutions of good quality in various areas of three-dimensional space formed by the dimensionality of problem N and technological limitations X max , W max of the LMC basis due to the presence of zone dependence. Combining the best aspects of ‘greedy’ and random approaches, it is possible to develop of a method that distributes the vertices in the blocks of the partition based on the weighting heuristics with aт adjustable degree of dispersion D relative to a greedy evaluation of the increment of solution quality. Special software was designed for this method; using this software a series of computational experiments has been carried out. In the course of matho-optimization, it was found that the optimal degree of dispersion D*=zero, which differs from the behaviour of the method in other problems of discrete combinatorial optimization and provides the iterative nature of the method only due to variation of the order of consideration of vertices. A computational experiment was carried out for the software implementation of the method using the indicated degree of dispersion. The experiment showed the advantage in terms of the quality of the output solutions for all quality parameters except for the intensity interblock interactions. At the same time, the working time of the method is 43 times more than that of the method of random selection and is limited by the time of evaluation of the quality of the generated partitioning.

To increase the accuracy of results, two methods for investigating wetting and spreading processes on a solid substrate are studied. In the first method, liquid drops are used; in the second method, bubbles are used. The shapes of drops and bubbles vary upon spreading, and this fact can be evaluated quantitatively only by means of the Laplace equation, but the equation is applied only in the case of bubbles (the second method). In the case of the first method, this circumstance excludes the purity control of the spreading drop surface. The influence of microcontaminations on the results is considered based on precision calculations performed for both methods. Spreading curves of nanobubbles with initial diameters of 20 and 10 nm on the substrates with various wettabilities are calculated, and wettability is evaluated not by the numerical value of the wetting angle but by the corresponding easily implemented examples of such substrate Γ, Φ, and H x , where x is the surface fraction under a bubble coated by ionogenic collector molecules: 0.8, 0.6, 0.4, and 0.2. Spreading curves clearly illustrate the range of possible nanobubble spreading from the limiting one on the Γ substrate to the almost zero one on the Φ substrate, as well as the sources of energy provision of spreading and their depletion causes. The informativity of spreading curves is caused by the fact that more than ten bubble and substrate parameters are applied in their calculation. When using the reagents, the flotation activation can spread to larger bubbles.

In the article was given statement of a problem of matrix multiplication. Is is show that desired problem can be simpl formulated but for its solving may be required both heuristic methods and set of algorithmic modifications relating to algorithmic and high-level software optimization taking into account the particular problem and allow to increase the multiplication performance. These include: a comparative analysis of the performance of the actions performed without GPU-specific optimizations and with optimizations, which showed that computations without optimizing the work with global GPU memory have low processing performance. Optimizing data distribution in global and local memory The GPU allows you to reuse the calculation time and increase real performance. To compare the performance of the developed software implementations for OpenGL and CUDA technologies, identical calculations on identical GPUs were performed, which showed higher real performance when using CUDA cores. Specific values of generation performance measured for multi-threaded software implementation on GPU are given for all of described optimizations. It is shown that the most effective approach is based on the method we can get much more performance by technique of caching sub-blocks of the matrices (tiles) in the GPU's on-chip local memory, that with specialized software implementation is provide the performance of 275,3 GFLOP/s for GPU GeForce GTX 960M.

The article covers a specific property of nanobubbles – spontaneous spreading over a solid hydrophobic substrate-particle adhered to them due to high capillary gas pressure in nanobubbles (Pk > 106 N/m2 ). The calculation principle of bubble spreading curves was considered and the parameter X responsible for intensity was introduced. The relation X(а) (а – bubble base diameter) was represented by a bimodal curve confirming that the process of nanobubble spreading was energetically supported by two independent sources acting sequentially. The first source was conditioned by reduction (approximately by 11 %) of nanobubble curvilinear surface area at the initial stage of spreading, and the second one was conditioned by the work of gas expansion due to a drop in Pk during the bubble spreading. The parameter X value was characterized by a significantly larger slope of the relation X(а) at the initial stage of spreading as compared to the second stage. As it turned out, the property which was found to determine the efficiency of industrial flotation processes in the past, now regains its application perspective. Due to the fact that this property becomes apparent in a limited range of bubble sizes, it was suggested to be labeled as a physical (or natural) fractal by analogy with Brownian motion which becomes apparent in a certain range of particle sizes. The influence of flotation agent surface activity on the shape of bubble spreading curves was demonstrated.

The specific property of nanobubbles with spontaneous spreading over the solid hydrophobic particle substrate adhered to them, which is caused by a high capillary gas pressure in nanobubbles ( P c > 10 6 N/m 2 ), is considered. The computational principle of bubble spreading curves is considered and parameter X characterizing the intensity is introduced. Dependence X ( a ) ( a is the bubble base diameter) is presented by a bimodal curve, which confirms that the nanobubble spreading is energetically provided by two sequentially acting independent sources. The first source is conditioned by the reduction (approximately by 11%) of the nanobubble curvilinear surface area at the initial spreading stage, and the second source is conditioned by the work of gas expansion caused by the drop of P c when the bubble is spreading. Parameter X is characterized by a considerably larger slope of dependence X ( a ) at the first spreading stage compared to the second one. It now turned out that the revealed property, which determines the efficiency of industrial flotation processes in past, finds prospects for application again after its recognition. Since it manifests itself in a limited range of bubble sizes, it is proposed to attribute it to the proper or natural fractal by analogy with the Brownian motion, which manifests itself in a definite range of particle sizes. The influence of the surface activity of flotation reagents on the shape of bubble spreading curves is shown.

In this paper we consider the problem of enumerating diagonal Latin squares of small order. In particular we discuss possible algorithmic approaches to this problem and show our results in this regard. Surprisingly, our research showed that the best algorithm for enumerating diagonal Latin squares consists of a number of fixed loops, and its effectiveness can be significantly increased by careful tuning and applying special heuristics. We used the constructed algorithm to enumerate all diagonal Latin squares of order 8. Also, it is being used to carry out large-scale computational experiment aimed at enumeration of diagonal Latin squares of order 9.

Automation of the cutting-speed control on lathe CNC equipment allowing one in real time to compensate for temperature deformations that adversely affect the accuracy of mechanical processing of the products is considered in the article. The analysis of the system control methods with using hard and soft arithmetic operations is given.

Abstract: The article describes the system of binary relations for graph-schemes of parallel algorithms, which are a special type of graphs with some special properties. Given a generalized parallel algorithm for building a matrix of binary relations based on private algorithms determine the connection, follow, alternative and parallelism relations. It is shown that using of specialized storage devices with a matrix structure can achieve a substantial gain in time during building the matrix of relations. In this case, the existing hardware with parallel structure (eg, GPU) are characterized by lower gets the win in time.

An algorithm, which is developed by the authors for adaptive interpolation, is based on analysis of brightness differences in the image, and allows maintaining clear boundaries in the image and excluding image blurring effect and a block defect, is discussed.