No 4 (2022)

The paper presents the multi-criteria choice problem with a numerical vector function on a subset of the vectors. It is assumed that the decision maker uses a fuzzy preference relation in the selection process. Information about the preference relation is considered to be known in the form of a finite collection of fuzzy quanta. We formulate an algorithm to reduce the Pareto set in the multicriteria choice problem using the set of quanta, and facilitate the final choice. A numerical example illustrates the algorithm work.

A method of solution of antagonistic game in violation of "common knowledge" principle is described, when players have incomplete knowledge about possible solutions and appropriate outcomes of the opposite side. As a formal model of a game situation it is proposed to use a fuzzy-multiple representation of estimates of possibilities of using by players their strategies and the corresponding consequences. The solution of this problem is based on the transformation of fuzzy estimates of the results of possible solutions for each situation in the form of an equivalent fuzzy set with a triangular identity function. The developed method does not impose restrictions on the affiliation functions of the initial fuzzy data. In addition to selecting the best solution, an estimation of its result and the degree of feasibility is obtained.

The paper considers the means for conceptual design of complex technical systems. We constructed a quasi-axiomatic theory that formalizes the procedures of generating meaning for a natural language description of the process of creating a new technical solution. Semantic categories, structures of universal sets, operations for comparing elements of the universe are introduced. The types of connection of elementary subsystems are described. We proposed a formalization of the procedure for multilevel synthesis of a technical system using a generative grammar over fuzzy structures. An example of designing a technical device is given.

The main methods for predicting traffic in telecommunications networks are the methods and techniques of Machine Learning (ML). As a rule, within the framework of the MO approach, the network traffic predictor is considered as a tool that uses one way or other accumulated statistics over time to draw conclusions about the future behavior of network traffic [4]. However, as the analysis of the literature shows, many modern MO tools, primarily neural networks, do not work efficiently enough due to the pronounced non-linearity of traffic changes and non-stationarity. Among the tasks of forecasting, the task of predicting signs of increments (direction of change) of the process of time series is singled out separately. The article proposes to use some results of the theory of random processes for a quick assessment of the predictability of signs of increments with acceptable accuracy. The proposed fast prediction procedure is a simple heuristic rule for predicting the increment of two adjacent values of a random sequence. Knowledge of the laws of time series probability distributions is not required. The connection with this approach for time series with known approaches for predicting binary sequences is shown. The possibility of using the experience of predicting the absolute values of traffic when predicting the sign of the change is also considered.

In the article, the authors present the results of the development and research of methods for creating 3D models of plants grown in vitro, which provide the ability to accurately record the morphometric indicators of the growth of individual parts, organs of plants and plants as a whole, cultivated on different nutrient media. The presented methods and algorithms in a complex solve the problems arising in the process of studying plants in a test tube, related to the complexity of the plant structure, the occurrence of distortions at the borders of the test tube, its possible fogging, as well as the influence of the human factor. A bank of 792 3D models for plants of 6 species has been created, which allows conducting simulation experiments to identify cause-and-effect relationships, forecasting and gaining new knowledge. The developed methods were checked for adequacy, an example of use for a specific plant was presented. The presented methods and algorithms can become the basis for the implementation of the process of digital phenotyping of plants.

Modern companion robots can solve a wide range of tasks while working together with a person. During the collaboration, a robot can receive commands from a person through various control systems, as well as use natural language. Utterances in natural language have significant degree of ambiguity (homonymy). In this paper we examine the methods, used to process utterances, and solve the possible homonymy during speech control of a robot in a natural or virtual environment.

The major problems associated with the rapid detection of forest fire and smoke are considered based on the use of autonomous unmanned aerial vehicles for flying over and surveying given areas. A method of searching for forest fires through a locally optimal flight route under conditions of uncertainty has been developed. We proposed original provisions of fuzzy sets, which allow an autonomous unmanned aerial vehicle to build an effective information-analytical model of situational-command control for moving along a route obtained in real time. A model of knowledge representation and processing have been developed that allow, on its basis, to automatically synthesize

logical-transformational rules for situational-command motion control of the aircraft's motion. It is shown that the proposed principle of building an information-analytical model makes it possible to reduce the complexity of selecting effective commands by significantly reducing the number of comparisons of the current situation with reference situations in the process of deriving solutions.