Journal archive

Both in Russia and in other countries, scientists note the need development of artificial intelligence methods for precision farming. However, the software nowadays used for this purpose is only developing and sometimes not released for sale, and the methods stated in individual articles use standard approaches, which leads to significant errors. Another reason is a non-trivial way of obtaining initial data, in particular, as a result of the analysis of satellite images, which is a rather laborious work and is possible only if there is a database of the objects under study with their specified geographical coordinates. The article discusses the development of a system for predicting the yield of agricultural land using satellite images. The system includes both classical methods (for example, the parabolic regression discussed in the article) and approaches developed by the authors to identify the parameters of the quasi-linear equation of n-factor autoregression. Despite the presence on the market of a fairly large number of software products for precision farming, many of the existing software and approaches are not intended to provide data for further analysis by specialists who are not employees of the developer company. As for our system, the collection and subsequent analysis of aerial photography data is carried out, which consists in building a model of the dynamics of the vegetation index, which is subsequently used to predict the yield of the field in the following periods. The proposed system is extensible, i. e., it allows connecting additional modules for analysis and inclusion in the analysis of additional factors that affect the indicator under study. The article describes the data presentation method, the process of calculating vegetation indices for a certain date, and also provides mathematical approaches to modeling the vegetation process using high-quality deterministic mathematical models and predicting the next season’s crop yield and production costs.

Modern economy is characterized by mass production and introduction of production services within the digital ecosystems that assist the customer in obtaining the final results, in particular, organization of procurement of radio-electronic components. The relevance of the task of optimal inventory rating control is related to development of a service that minimizes risks arising from uncertainty of such factors as receipt of requests for commodity and its supply, supplier’s prices and selling prices, etc. The issues of development of economic and mathematical models of inventory rating optimization applicable to the specific features of vertically integrated companies and legally independent organizations operating in conditions of the market economy on kanban principles are considered. It is shown that in conditions of uncertainty, use of ratings allows to increase the efficiency of the inventory control system. The possibility of presentation of the inventory control system in the form of an evolutionary simulation model is substantiated. The procedure is provided to bring the task of inventory rating to the canonical form of an evolutionary simulation model (ESM). The peculiarities of the inventory control systems are revealed using the examples of “Stock” ESM for VIC and “Zapas” ESM for independent companies. The methods are determined for calculation of the derived rates: the average number of supplies, the average interval between supplies, etc. The contribution of the authors of the article consists in development of the ESM applicable to inventory rating, which combines the simulation models (estimates of the actual overstatement and understatement cost options relative to the planned radioelectronic components inventory rate, as well as estimates of the expected costs (risks) obtained in statistical tests) with the optimization models implemented based on the theory of equilibrium random processes. The aggregate strategy, i.e. the strategy aimed at minimizing the amount of the overstatement and understatement risks is used as a criterion for inventory rating. The experimental studies of “Stock” and “Zapas” models implemented using the “Equilibrium” module of the author’s system – “Decision” allow real-time modeling of various options of the studied rates.

Digital counterparts of business processes in modern conditions are used not only to analyze the process, but also to optimize its execution. The presence of feedback from the real process makes it possible to create effective automated solutions for the execution of the constituent elements of the process. In this scientific study, an improved mechanism for managing elements of an instance of a real business process using its digital counterpart is proposed. The proposed mechanism is based on the implementation of software for the orchestration of the API component call in the workflow and the orchestration of RPA robots in the logical and structural layer of the virtual representation of the digital twin of the business process. RPA robots simulate human actions on a computer by interacting with various software applications, processing data and performing operations. Orchestration, in turn, allows the orchestrator, a software or system component, to manage the sequence and flow of tasks, resources and data within the process. Since in practice, business entities, among others, solve the problem of developing and applying digital twins of reference and original business processes, this study also presents a conceptual scheme for developing a digital twin of a business process. The issues of designing a virtual business process model, at the cyber-physical level, the level of extraction and preprocessing, the level of models and algorithms, and the level of visual representation of the digital counterpart of the business process are also investigated. The benefits of running a business process using its digital counterpart include increased error reduction, faster process execution, improved quality, and the ability to easily make changes to the process in response to changing requirements. This is especially useful in modern organizations where efficiency and flexibility in managing business processes are required.

A computer model and test results of a decentralized self-controlling UAV system using the molecular dynamics method and various forms of interaction potential are presented. In this model, each UAV is replaced by a “quasi-molecule” with point dimensions and a mass equal to the mass of the UAV. The motion of these “quasi-molecules” is calculated using Newtonian mechanics, taking into account the potential force field from other “quasi-molecules”. The system consists of a group of UAVs performing various ordered formations in accordance with a given mission, as well as obstacles and “hunters” trying to collide with the UAV. “Hunters” are also represented by point-like “quasi-molecules”. This allows us to simulate the presence in the system of both active and passive interference with UAV movement. When modeling passive interference, “hunters” are located on the surface of an obstacle at the nodes of a surface mesh superimposed on it. When modeling the movement of UAVs, taking into account their interaction with each other, active and passive interference, specially designed interaction potentials were used. Numerous tests have been carried out on the system under these conditions. No collisions of the UAVs with each other, with obstacles or with “hunters” were observed. This confirms the reliability of the developed principles for controlling the movement of a group of UAVs under conditions of intense active and passive interference. The main difference between the presented work and existing ones is that various missions are carried out by groups of UAVs without external control according to data from a technical 3D vision system within the framework of a fairly simple molecular dynamics model by changing the shape of interaction potentials.

One of the most important components of the global Internet are traffic control and management systems. In order to achieve uninterrupted information and communication interaction, the organization of the process is constantly changing, covering not only individual subnets, but also p2p network architectures. The dominant areas for improving the network structure include 5G, IoT and SDN technologies, but their implementation in practice leaves the issue of ensuring the information security of networks built on their basis without a satisfactory solution. Current virtual tunnel deployment topologies and intelligent traffic distribution components provide only partial solutions, particularly in the form of access control based on user traffic and security through dedicated user certificates. The deployment of a tunnel is of particular importance in cases where it is necessary to ensure consistency and coordination of the work of complex socio-economic systems, an example of which is the information and communication exchange between participants in scientific and industrial clusters formed to implement projects for the creation of innovative products. However, existing solutions have disadvantages such as the need to purchase a license for full-featured access to the software product and specialized configuration of client-server authentication that provides secure access to a remote network route. The approach proposed by the authors, based on neural network distribution of traffic between clients of a private dedicated network, allows us to eliminate the noted shortcomings. Based on this principle, a multi-module system for intelligent packet routing was created and tested through unit testing. An analysis of the effectiveness of using a trained network address distribution model is presented in comparison with the use of a DHCP server based on the isc-dhcp-server package, distributed as the dhcpd service.

One of the important social tasks is the task of improving the efficiency of transport service systems. The paper proposes approaches to solving the problem of increasing the efficiency of contactless fare payment on passenger road transport, which are based on geolocation technology. Based on the consideration of fare payment schemes for closed and open cycles, scenarios for the implementation of fare payment using a flat fare system, a zonal fare system and electronic tickets with QR codes are presented. Several payment implementation work schemes have been developed based on the application of the geolocation concept and using various technologies to implement this concept, namely 1) Bluetooth Low Energy, 2) obtaining coordinates from GPS/GLONASS navigation systems, 3) data received through Wi-Fi access points and 4) location data provided by basic stations of GSM networks. A comparative analysis was carried out for each proposed option, a basic set of technical and software support tools was determined, and recommendations were given on the conditions that determine the feasibility of its application. For experimental confirmation of the operability of each of the options for obtaining the estimated values of the main characteristics, operating samples were constructed on which the corresponding measurements were carried out.

The article shows the problem of reducing the efficiency and quality of the formation of control actions by the head of the class when preparing groups of specialists to perform complex tasks agreed on time, place and goals in organizational and technical systems of various purposes due to the influence on the psychophysiological capabilities of the head of the class of increasing the density of the flow of incoming information at the stages of working out the interaction between groups of specialists in a single virtual space. The use of intelligent methods to solve this problem is substantiated and a method of scenario-information analysis of adaptive training of groups of specialists is proposed, the essence of which is to build a scenario-information model of adaptive training based on a fuzzy ontological approach, taking into account the availability of resources, the current state of specialists (current level of preparedness) and precedent scenarios of adaptive training for subsequent modeling of this process and estimates of its achievability under various preparation scenarios. A fuzzy ontological approach to modeling adaptive training of groups of specialists using granulation of information resources is proposed, which makes it possible to organize the training process more efficiently, including in conditions of limited time and material resources. The results of experimental studies on improving the level of preparedness of group specialists through intelligent management of their adaptive training are shown.

Studies of the influence of the chemical composition of ore raw materials on the properties of foundry products, as a rule, do not consider the features of production processes. The main reasons for this are that, firstly, the results of such research are usually limited to laboratory studies, secondly, the models do not take into account the specifics of the structure of ore raw materials, since these data can be obtained only after the completion of all thermophysical and chemical-energy-technological processes of heat treatment, thirdly, the use of models is difficult due to the need to promptly account for changes in the composition and values of systemic and external factors during heat treatment. The purpose of the paper is to design neural network models and tools that provide the possibility of adaptive structural and parametric adjustment to changes in the parameters of the analyzed processes of heat treatment of ore raw materials. Data on the chemical composition of ore raw materials and information on heat treatment processes are used as parameters of the projected neural network models. As a result of the research, the hypothesis about the possibility of indirect accounting for the influence of structural features of ore raw materials on the quality of products has been confirmed. Confirmation of this hypothesis will allow us to offer effective tools for operational management of thermophysical and chemical-energy-technological processes of heat treatment of ore raw materials.

A method for solving the inverse kinematics problem for a three-link robotic manipulator is proposed based on one of the types of machine learning - reinforcement learning. In the general case, this task consists of finding the laws of change in the generalized coordinates of the manipulator’s gripping device that provide the specified kinematic parameters. When solving the problem analytically, the basis for calculating inverse kinematics is the Denavit – Hartenberg parameters with further implementation of numerical matrix calculations. However, taking into account the kinematic redundancy of multi-link manipulators, this approach is labor-intensive and does not allow automated consideration of changes in the external environment in real time, as well as the features of the robot’s field of application. Therefore, an urgent research task is to develop a solution whose structure contains a self-learning block that provides a solution to the inverse kinematics problem under conditions of a changing external environment, the behavior of which is unknown in advance. The proposed method is based on simulating the process of achieving the goal of robot control (positioning the gripping device of the manipulator) at a given point in space using the trial and error method. For approaching the goal at each learning step, a reward function is calculated, which is used when controlling the robot. In the proposed method, the agent is a recurrent artificial neural network, and the environment, the state of which is observed and assessed, is a robotic manipulator. The use of a recurrent neural network made it possible to take into account the history of the movement of the manipulator and overcome the difficulties associated with the fact that different combinations of angles between links can lead to the same point in the workspace. Testing of the proposed method was carried out on a virtual model of the robot, made using the MatLAB Robotics System Toolbox and the Simscape environment, which showed high efficiency in terms of the “time – accuracy” criterion of the proposed method for solving the inverse kinematics problem.