Irina Vatamaniuk

This study addresses the foundational step of developing a classification and taxonomy of agent objective functions as a prerequisite for analyzing stability and forming robust production schedules in distributed manufacturing systems. The research is based on the premise that instability or insufficient robustness in scheduling solutions often arises from the neglect of the inherent multi-agent nature of real-world distributed production systems. These systems are characterized by the presence of multiple decision-making entities, each pursuing its own objectives or performance indicators. Since strategic management in such systems is typically oriented toward achieving global system-level goals, it often overlooks the interests of individual agents. As a result, the implemented decisions may encounter resistance from specific agents and lead to deterioration in the performance of their individual objective functions. These features underline the need to develop tools for identifying robust solutions, in which both the system as a whole and its constituent agents can achieve sustainably high performance across their respective objectives. The aim of this study is to analyze the divergent objective functions of management agents in distributed manufacturing systems in the context of forming robust production schedules. The research explores typical objective functions of structural units within the production system and presents their classification in terms of constraints, nature, granularity, behavioral orientation, and inter-agent dependency. The outcomes of the study include a comprehensive taxonomy of agent objective functions, along with the selection of relevant game-theoretic models for each pair of agents based on their interaction strategies. The findings contribute to the development of methodological and technological tools for decision support in sustainable manufacturing, extending current research on intelligent agent modeling and coordination in complex production environments.

Typically, the quality of the bitumen adhesion in asphalt mixtures is assessed manually by a group of experts who assign subjective ratings to the thickness of the residual bitumen coating on the gravel samples. To automate this process, we propose a hardware and software system for visual assessment of bituminous coating quality, which provides the results both in the form of a discrete estimate compatible with the expert one, and in a more general percentage for a set of samples. The developed methodology ensures static conditions of image capturing, insensitive to external circumstances. This is achieved by using a hardware construction designed to provide capturing the samples at eight different illumination angles. As a result, a generalized image is obtained, in which the effect of highlights and shadows is eliminated. After preprocessing, each gravel sample independently undergoes surface semantic segmentation procedure. Two most relevant approaches of semantic image segmentation were considered: gradient boosting and U-Net architecture. These approaches were compared by both stone surface segmentation accuracy, where they showed the same 77% result and the effectiveness in determining a discrete estimate. Gradient boosting showed an accuracy 2% higher than the U-Net for it and was thereby chosen as the main model when developing the prototype. According to the test results, the evaluation of the algorithm in 75% of cases completely coincided with the expert one, and it had a slight deviation from it in another 22% of cases. The developed solution allows for standardizing the data obtained and contributes to the creation of an interlaboratory digital research database.

One of the promising areas of using robotic systems (RS) is group control of robots. The means of ensuring the operator's interaction with autonomous RS for monitoring and remote control - the human-machine interface (HMI) of the robotic systems, are discussed in the paper. The foreign and domestic publications describing existing approaches and priority directions to the construction of HMIs for various types of robotic systems - ground and air, as well as modern commercial solutions in this area are analyzed in the paper. The performed analysis allows us to conclude that the most popular and convenient type of human-machine interface for the types of robotic systems under consideration today is a digital application with a graphical interface. To implement complex scenarios of human-machine interaction, combinations of several methods are used, such as: a digital application with a graphical interface, virtual reality tools, gesture recognition, voice commands. The technical appearance of the man-machine interface of group unmanned robots based on the results of the analysis was developed. It ensures the prompt formulation of the applied problem to be solved and the subsequent control of its execution by the operator. The proposed solution is intuitive and flexible, which provides the user with a higher speed of information processing. Key words Group robotics, human-machine interface (HMI), group control of robots, HMI architecture, robotic system.

Purpose of research.Emotions play one of the key roles in the regulation of human behaviour. Solving the problem of automatic recognition of emotions makes it possible to increase the effectiveness of operation of a whole range of digital systems such as security systems, human-machine interfaces, e-commerce systems, etc. At the same time, the low efficiency of modern approaches to recognizing emotions in speech can be noted. This work studies automatic recognition of emotions in speech applying machine learning methods.Methods.The article describes and tests an approach to automatic recognition of emotions in speech based on multitask learning of deep convolution neural networks of AlexNet and VGG architectures using automatic selection of the weight coefficients for each task when calculating the final loss value during learning. All the models were trained on a sample of the IEMOCAP dataset with four emotional categories of ‘anger’, ‘happiness’, ‘neutral emotion’, ‘sadness’. The log-mel spectrograms of statements processed by a specialized algorithm are used as input data.Results.The considered models were tested on the basis of numerical metrics: the share of correctly recognized instances, accuracy, completeness, f-measure. For all of the above metrics, an improvement in the quality of emotion recognition by the proposed model was obtained in comparison with the two basic single-task models as well as with known solutions. This result is achieved through the use of automatic weighting of the values of the loss functions from individual tasks when forming the final value of the error in the learning process.Conclusion.The resulting improvement in the quality of emotion recognition in comparison with the known solutions confirms the feasibility of applying multitask learning to increase the accuracy of emotion recognition models. The developed approach makes it possible to achieve a uniform and simultaneous reduction of errors of individual tasks, and is used in the field of emotions recognition in speech for the first time.

To extend the application of the late waves of the event-related potentials (ERPs) to multiple modalities, devices and software the underlying physiological mechanisms and responses of the brain for a particular sensory system and mental function must be carefully examined. The objective of this study was aimed to study the sensory processes of the “human-computer interaction” model when classifying visual images with an incomplete set of signs based on the analysis of early, middle, late and slow ERPs components. 26 healthy subjects (men) aged 20–26 years were investigated. ERPs in 19 monopolar sites according to the 10/20 system were recorded. Discriminant and factor analyzes (BMDP Statistical Software) were applied. The component N₄₅₀ is the most specialized indicator of the perception of unrecognizable (oddball) visual images. The amplitude of the ultra-late components N₇₅₀ and N₉₀₀ is also higher under conditions of presentation of the oddball image, regardless of the location of the registration points. In brain pathology along with the pronounced asymmetry of the wave distribution, reduction of the N₁₅₀ wave and lengthening of its peak latency, a line of regularities were noted. These include–a pronounced reduction in peak latency P₂₅₀ and N₃₅₀, an increased amplitude of N₃₅₀ in the frontal and central points of registration, a decrease in the amplitude of N₄₅₀ in the left frontal cortex and its increase in the occipital registration points, activation of the occipital cortex at a time interval of 400–500 ms, as well as fusion later waves. We called such phenomena of the development of cognitive ERP in brain pathology “the incongruence of ERP components”. The results of the research are discussed in the light of the paradigm of the P₃₀₀ wave application in brain-computer interface systems, as well as with the peculiarities in brain pathology.