The current level of development of unmanned transport aviation makes it possible to use it widely for the automation and robotization of many processes in agriculture. One of the promising applications of unmanned aerial vehicles is the delivery of essential supplies to remote areas cut off from land transportation routes. The urgent task here is the creation of mobile solutions that enable to quickly organize transportation of cargo, including heterogeneous ones, under conditions of shortage of certain types of cargo and the need to prioritize requests for delivery. The paper proposes a model for creating an optimal transportation plan of heterogeneous cargo between sources and destinations for the route network of arbitrary structure. The model contains mechanisms for formalizing and taking into account the priority of cargo delivery tasks for the different types of priority, as well as a mechanism for balancing the priority of delivery tasks and the transportation cost. The results of testing the proposed optimization model for planning the process of cargo delivery to remote areas, with visualization of the transportation plan based on a digital terrain model are presented. The directions for further development and improvement of the proposed approach are discussed.

The task of creating a neural interface for controlling a robotic system by means of an oculographic interface and bioelectric signals, is considered. The article highlights the results of scientific experimental research aimed at the evaluation of the representativeness of bioelectrical signals obtained by electroencephalography (EEG). The basic hypothesis is formulated and tested with the help of artificial neural network technology. The authors consider an experiment on the formation of steady-state visually evoked potentials in a group of people with the subsequent creation of an applied database. They describe an original approach for extracting representative features from the EEG signal. With the help of deep machine learning technology the representativeness of the data under study is evaluated. The main conclusions are formulated and the hypothesis that each brain lead reproduces unique waves which are characteristic of each brain zone is confirmed. The proposed model of a symmetric multilayer multi-adaptive direct propagation neuron can find its application in solving problems related to the processing of EEG signals. Based on the results of this study, the authors suggest that data on the bioelectrical activity of the brain can be uniquely identified, and thus used as control signals for various robotic devices.

Conservation and increase of fish resources is one of the priority tasks of modern society. Due to agricultural activities, the impact of the construction of hydroelectric power stations, as well as the widespread presence of poachers have led to the fact that fish resources need to be replenished in special fisheries. The article points out the importance and conservation of fish resources around the world, as well as the priority of caviar counting in the breeding of sturgeon species. Conducting such activities on the farms is necessary to monitor the impact and effectiveness of the use of various feeds, changing conditions, etc. Traditionally, two methods are used at enterprises to count caviar—by weight and volume, in addition, the von Bayer method with the use of a special tool is used in practice to count red caviar. The paper also considers high-performance digital methods of caviar counting, outstanding and requiring technical solutions to perform a design for automatic counting using neural network algorithms. An original design using a camera with Internet access is proposed. Results of experimental systems for sturgeon fish caviar reproduction are presented.

The problems of scattering of electromagnetic waves by anisotropic dielectric structures are studied on the basis of volumetric singular integral equations. Existence and uniqueness theorems for solutions are given for a wide class of media, including lossless media. An efficient method for solving integral equations based on the collocation method and fast Fourier transform algorithms is described.

Parallel and distributed computing have become widespread as one of the main ways to improve the performance of computing environments when processing big data. The issue of assessing the main performance indicators of such systems remains relevant to the present time due to the complexity of such analysis. A fork-join queuing system is well suited for modeling parallel data processing. The article explores the possibility of estimating the average response time of the system using an analytical expression for the upper bound of the maximum order statistics. Two options are described, depending on the amount of knowledge available about the operation of a high-performance environment. The results of the numerical experiment confirm the efficiency of the approach for a fairly wide range of parameters, one of the main advantages of which is the ease of use.