Кафедра комп’ютерно-інтегрованих технологій, автоматизації та робототехніки (КІТАР)
Постійний URI для цієї колекції
Перегляд
Останні подання
Публікація Methods of Automated Monitoring and Control System of Greenhouse Complex(2025) Dolhosheia, I.; Tsymbal, O.The study considers methods of an automated monitoring and control system for a greenhouse complex aimed at increasing production efficiency and optimizing resource consumption. The proposed mathematical models describe the dynamics of the main microclimate parameters, in particular temperature, humidity, CO₂ concentration and soil moisture, which allows for precise process control. Based on the obtained models, a decision-making logic was built that combines threshold rules, fuzzy logic and adaptive control, ensuring flexibility and reliability in changing conditions. The numerical simulation demonstrated the system's ability to maintain stable environmental parameters with minimizing energy costs and water consumption. The results confirm that the implementation of such systems contributes to increasing yields, improving product quality and sustainable development of the agricultural sector in the conditions of Industry 4.0 and Industry 5.0.Публікація Numerical Study of Algorithms to Construct Optimal Trajectories for Collaborative Robots in Industry 5.0 Manufacturing Scenarios(2025) Elgun JabrayilzadeThis study considers the problem of numerical modeling of algorithms for constructing optimal trajectories for collaborative robots in Industry 5.0 production scenarios, where the key aspect is safe and effective interaction with a dynamic environment and humans. The developed mathematical models are based on potential function methods and multi-criteria optimization, which allows combining obstacle avoidance, energy consumption reduction, and task execution time minimization. Numerical modeling has shown that the actual trajectory of the robot remains convergent to the desired one even in the presence of obstacles, confirming the adaptability of the algorithms to unpredictable changes in the environment. Analysis of the time error showed its short-term increase during maneuvering and subsequent stabilization, which indicates the reliability of the proposed models. The study of the smoothness of movement confirmed the absence of sharp changes in the curvature of the trajectory, which ensures safety and comfort of operation in production processes. The results obtained prove the feasibility of using the developed approaches to create intelligent robotic systems capable of increasing production efficiency, reducing accident risks, and improving the quality of interaction between robots and humans in the Industry 5.0 concept.Публікація Analysis of operator identification methods in the working area of a collaborative manipulator robot(2025) Molozhanov, L.; Gurin, D.The paper presents the results of the analysis of modern methods of operator identification in the working area of a collaborative robot-manipulator, which is important in the context of safe and effective interaction between a person and a robotic system. Sensor approaches, computer vision methods, biometric technologies, algorithms based on deep neural networks, as well as integration solutions using data fusion and fuzzy logic are considered. It is shown that sensor methods provide basic security, but are limited in accuracy, while computer vision and deep learning allow achieving high detail in identification, although they require significant computing resources. Biometric approaches create conditions for personalization of interaction, but may lose effectiveness in a production environment with a high level of noise and the need to use personal protective equipment. Data integration methods increase the stability of the system and its ability to work in conditions of uncertainty, ensuring multi-level adaptability. The analysis confirms that optimal solutions should be based on a combination of several methods, which is consistent with the concept of Industry 5.0 and contributes to the development of new generation cognitive robotic systems.Публікація Застосування інтелектуальних систем управління робототехнічними системами для досягнення цілей сталого розвитку у сфері гуманітарного розмінування(2025) Янушкевич Д. А.; Іванов, Л.; Толкунов, І.У доповіді розглядаються актуальні питання щодо досягнення цілей сталого розвитку у сфері гуманітарного розмінування із застосуванням робототехнічних систем. Розв’язання цієї проблеми потребує креативності, комплексного підходу та інтелектуальних систем управління. Це дозволяє розробити моделі управління робототехнічною системою як на рівні прийняття рішень, так і на виконавчому рівні.Публікація Using digital twins and artificial intelligence for the synchronization of physical and virtual collaborative robots(2025) Lisovskyi, A.This study proposes an approach to synchronizing physical and virtual collaborative robots based on the concept of digital twins and artificial intelligence tools. The proposed mathematical models allow formalising the processes of reflecting the real state of robots in a digital environment and minimising synchronisation errors. Particular attention is paid to the use of prediction, data filtering, and reinforcement learning algorithms that ensure the adaptability and stability of the system. The paper analyses the advantages of direct, predictive, and hybrid synchronisation methods and evaluates their effectiveness in a multi-user environment. The use of artificial intelligence allows for an increase in the level of autonomy and safety of human-robot interaction. The results of the study demonstrate the promise of integrating digital twins into modern robotic systems and open up opportunities for creating scalable and flexible manufacturing solutions.Публікація Development of a model for decentralized control of a group of collaborative robot manipulators(2025) Maksymova, S.; Shakhov, P.The paper considers an approach to developing a model for decentralized control of a group of collaborative robot manipulators, focused on increasing autonomy, adaptability, and safety in a dynamic production environment. The proposed solution allows minimizing dependence on centralized computing resources, reducing the risks of system failures, and ensuring effective interaction of robots in the joint performance of complex manipulation tasks. The results of the study demonstrate the prospects of decentralized models for implementing the concepts of Industry 5.0.Публікація Research on Methods for Controlling a Group of Mobile Robots Under Uncertainty(2025) Moisieiev, M.; Yevsieiv, V.The article considers modern methods for controlling a group of mobile robots in environments with a high level of uncertainty, which is a relevant task in the context of the development of the Industry 5.0 concept. The main attention is paid to the analysis of centralized and decentralized approaches, probabilistic models, fuzzy logic, collective intelligence methods, and reinforcement learning algorithms. The conducted research demonstrates their features, advantages, and limitations when applied in production and service scenarios. It is shown that decentralized and bioinspired methods provide high stability and scalability, while probabilistic models and fuzzy logic allow you to work effectively with incomplete information. Machine learning methods provide the ability to adapt and self-learning, but require significant computing resources. The results obtained indicate the feasibility of integrating different approaches into hybrid systems, which allows you to increase the efficiency and reliability of group control of robots in complex environments.Публікація Analysis of Object Identification Methods for FPV Drones(2025) Chebanchyk, D.; Yevsieiv, V.The abstracts of the report consider modern methods of object identification for FPV drones with an emphasis on their application in real time and under conditions of limited computing resources. Classical approaches based on keypoint extraction, deep convolutional neural networks, semantic and instance-segmentation methods, as well as state filters and lightweight optimized models are analyzed. The study shows that each of the methods has its advantages and limitations depending on the accuracy, processing speed and complexity of the environment. Special attention is paid to hybrid approaches that combine the advantages of several methods to ensure stable and effective object identification on board FPV drones. The results obtained emphasize the need to optimize algorithms and adapt models to the resource constraints of drones to ensure reliability and accuracy of operation in dynamic conditions.Публікація Використання методів комп’ютерного зору та штучного інтелекту для автоматизації підготовки CAD-документації друкованих плат(2025) Онищенко, В.; Малий, О.; Мірошніченко, В.Розглянуто застосування методів комп’ютерного зору та штучного інтелекту для автоматизації підготовки CAD-документації друкованих плат. Показано обмеження традиційних ручних підходів і сучасних CAD/CAE-систем у випадках відсутності вихідних проєктних файлів. Проаналізовано ключові етапи процесу — попередню обробку зображень, сегментацію, класифікацію та формування структурованих даних. Окреслено основні виклики та перспективи розвитку, зокрема створення відкритих датасетів і відновлення електричних схем.Публікація Оцінка технічного стану технологічного обладнання та діагностика неполадок у його роботі в умовах невизначеності(2025) Білоусов, М.; Стародубцев, М.; Шибанов, С.; Невлюдова, В.; Макаренко, Г.У статті показано, що проблема оцінки технічного стану технологічного обладнання в умовах невизначеності може розглядатися як проблема розпізнавання образів і розроблено класифікаційну модель оцінки технічного стану технологічного обладнання, що реалізує один з можливих підходів до розпізнавання стану об’єкта діагностування – виділення ознак та класифікацію з подальшою ідентифікацією стану об’єкта. Встановлено, що найчастіше аварійна ситуація в роботі технологічного обладнання характеризується неявно вираженими ознаками. Запропоновано модель діагностики неполадок в роботі обладнання, яка дозволяє в умовах неповноти інформації та обмеженості часу діагностування прийняти рішення про тип дефекту з множини можливихПублікація Integration of Artificial Intelligence in Assistive Robots: Challenges and Opportunities(2025) Stetsenko, K.Assistive robots are increasingly becoming an essential component in providing support to people with disabilities and elderly individuals. The integration of artificial intelligence (AI) enhances their capability to interact naturally with humans, understand context, and provide adaptive assistance. This article discusses current AI applications in assistive robotics, technical and ethical challenges, and opportunities for future development. The study also includes an overview of experimental mobile robotic platforms with manipulator capabilities.Публікація Digital Technologies for Monitoring the Dielectric Properties of Carbon-Carbon Composites(2025) Ovcharenko, V.; Tokarieva, O.The paper discusses the implementation of digital technologies in monitoring the dielectric properties of carbon-carbon composites. It is shown that real-time me asurement of complex dielectric permittivity enables control of key parameters during graphitization, pyrolysis, and impregnation processes. The integration of dielectric spectroscopy methods with automated control and digital modeling systems is substantiated, ensuring adaptability and accuracy of modern manufacturing processes.Публікація Comparative Analysis of Neural Network Architectures for Intelligent Microclimate Control in Production(2025) Yevsieiev, V.; Holod, I.A comparative analysis of neural network architectures (MLP, RNN, NNARX) for predicting microclimate parameters in industrial cyber-physical systems has been carried out. The advantages of NNARX in reproducing environmental dynamics are demonstrated, and its application for intelligent control is substantiated.Публікація Mathematical Model of Adaptive Control of a Collaborative Mobile Manipulator in a Shared Working Environment(2025) Yevsieiv, V.This paper presents a mathematical model of adaptive control of a collaborative mobile manipulator in a shared working environment that meets the requirements of safe and effective human-robot interaction. The developed approach is based on an adaptive controller that provides accurate tracking of the desired trajectory even in the presence of parameter uncertainties. Numerical simulations have shown that the actual robot trajectory practically coincides with the given one, and the error does not exceed 0.05 rad and decreases asymptotically over time. In addition, the parametric adaptation system demonstrated a gradual approximation of the estimated parameter to the true one, which confirms the model's ability to self-learn in dynamic conditions. The results obtained indicate the high efficiency of the proposed method, its stability and prospects for implementation in the production processes of Industry 5.0.Публікація Adaptive Regulation of the Manipulator's Movement Speed Depending on the Distance to the Person and the Level of Load on the Actuator(2024) Yevsieiev, V.; Maksymova, S.; Starykova, S.; Jafar AbabnehThe article presents an approach to adaptive control of the manipulator speed based on fuzzy logic, which allows taking into account the distance to the person and the level of load on the executive body. The proposed model forms control actions in accordance with the logic of safe human-robot interaction, providing a dynamic change in speed to increase efficiency and safety. The system uses a set of linguistic rules and triangular membership functions to process fuzzy input data. The simulation results demonstrate the stable behavior of the system and confirm the feasibility of using fuzzy approaches in controlling the movement of manipulators in variable conditions. The developed approach has high potential for implementation in robotic systems that operate in close proximity to a person. Development prospects include integration with machine learning and realtime use on embedded platforms.Публікація Development of A Method for Transmitting and Encoding Technical Information About the State of a Mobile Robot Based on Least Significant Bit(2024) Chala, O.; Yevsieiev, V.; Maksymova, S.; Ahmad AlkhalailehThe rapid development of Industry 5.0 emphasizes the need for effective communication between humans and robotic systems, particularly mobile robots operating in dynamic environments. Traditional data transmission methods face challenges related to latency, security, and adaptability, necessitating new approaches to ensure seamless information exchange. This paper presents a method for encoding and transmitting technical information about a mobile robot's state using computer vision, enhancing reliability and autonomy. The proposed approach reduces dependence on network infrastructure, increases data transmission efficiency, and ensures robustness under varying environmental conditions. The findings contribute to the advancement of smart manufacturing by integrating vision-based communication into industrial robotics.Публікація Adjusting the Movements of the Robotic Platform Through Inverse Kinematics(2025) Nevliudov, I. Sh.; Gurin, D.; Yevsieiev, V.The paper presents an approach to the development of a control system for a quadruped robotic platform using inverse kinematics. The solution enables accurate movement of the robot's legs to achieve stable motion and balancing under uncertain terrain conditions. The kinematic model is calculated using FreeCAD tools and implemented in a control code through the transformation of joint angles into PWM signals for servo drives. A simulation model in MATLAB Simulink was created to test the stabilization algorithm. The obtained results demonstrate the ability of the system to correct the robot's posture and maintain balance in real time, even under significant external disturbances .Публікація Features of the Development of a Humanoid Robot Control System on ESP8266MOD(12F)(2025) Yevsieiev, V.; Maksymova, S.The abstracts consider approaches to the development of a decentralized control system for a small-sized humanoid robot based on ESP8266MOD(12F) taking into account the requirements of Industry 5.0. A control architecture combining fuzzy logic, adaptive algorithms and inter-agent interaction to achieve individual and group goals within a robotic team is presented. Particular attention is paid to the mathematical description of the system's behavior, as well as the practical implementation of control in conditions of limited hardware resources. The work is aimed at the development of cognitive collaborative robotics in educational and scientific applications.Публікація Implementation of STEM education in distance learning conditions during martial law in Ukraine: challenges, tools and prospects for training future engineers(2025) Yevsieiev, V.; Starikova, S.The paper considers the relevance of implementing STEM education in the context of distance learning caused by martial law in Ukraine. The main challenges associated with the lack of access to physical laboratories, technical inequality and insufficient digital competence of teachers are analyzed. The use of virtual simulators, in particular Tinkercad, Fritzing, Proteus, is proposed as effective tools for the formation of engineering skills. A comparative analysis of these environments is conducted from the point of view of their educational potential. The prospects for integrating the STEM approach into school and university education are outlined, taking into account the crisis conditions and the needs for the restoration of educational infrastructure.Публікація Hybrid Approaches to Building Intelligent Robotic Systems on FPGAs and MCUs for Industry 5.0(2025) Yevsieiev, V.; Maksymova, S.; Demska, N.; Starodubcev, N.The paper considers hybrid approaches to the development of intelligent robotic systems using microcontrollers (MCU) and programmable logic integrated circuits (FPGAs) in the context of Industry 5.0 tasks. The advantages and limitations of existing technical solutions in the construction of cognitive and collaborative robots are analyzed. The feasibility of combining MCU and FPGAs within the framework of decentralized control systems capable of adaptation, interaction and learning in real time is substantiated. The proposed approaches demonstrate the potential for increasing the efficiency of distributed robot systems in human-centric technologies.