The dissertation is devoted to the development of methods and tools for planning the logistics of autonomous cyber-physical systems. The modern world is developing rapidly, especially in the field of information technologies. The use of artificial intelligence in cyber-physical systems is a key factor in this development. Dynamic changes require new approaches to management, and this is what makes the research topic extremely relevant. Autonomous logistics cyber-physical systems are becoming increasingly complex and extensive. Effective management of them becomes a challenge due to the variety of parameters affecting their functioning. The use of artificial intelligence methods can greatly facilitate this process and increase its effectiveness. Effective management of logistics cyber-physical systems is important for saving resources, time and money. The use of artificial intelligence technologies can help optimize processes, increase productivity and reduce costs. Autonomous systems are widely used in various industries, including transportation, manufacturing, medicine, and more. Increasing autonomy of such systems creates new challenges such as security, reliability and efficiency. The research dedicated to the solution of these problems is extremely important, as it allows to improve the functioning and implementation of autonomous logistics cyber-physical systems at the international level. The application of artificial intelligence in the management of cyber-physical systems requires a deep understanding of the integration of various technologies. Solving this problem not only contributes to the development of a specific direction, but also contributes to the general development of scientific thought and society as a whole.
Therefore, the relevance of the presented dissertation work is determined by the modern world's need for effective methods of managing complex autonomous logistics cyber-physical systems that use artificial intelligence technologies. The results of the research have a significant impact on the development of information technologies and the improvement of the functioning of important branches of economy and science.
The aim of the research is to improve the efficiency of real-time planning of the logistics of autonomous vehicles as a component of a cyber-physical system using reinforcement learning models.
The results obtained in the dissertation research contain scientific novelty:
1. a new method of path planning for autonomous logistics cyber-physical systems has been developed, which includes modifications of the A* algorithm, which provide control of the movement of autonomous vehicles in logistics cyber-physical systems, using reinforcement learning models, to improve decision-making in real time;
2. a model-oriented method for modeling cyber-physical systems has been improved, which, unlike the existing ones, involves a symbiosis of meta-modeling and business process models, which allows determining hierarchical structures, dependencies between various elements of cyber-physical systems, including aspects of their dynamics, resource management and interaction with the environment; and ensures the effectiveness of such systems and their compliance with strategic goals;
3. the method for building an adaptive logistics system based on the integration of artificial intelligence tools to optimize route planning and avoid obstacles was further developed, which, unlike existing ones, allows the logistics system to flexibly adapt to changes in internal and external conditions and avoid obstacles in time is determined by high performance and satisfaction of the requirements of the modern logistics environment, where the speed of response and effective management of resources determine the success of the system.
The practical significance of the results of the dissertation research lies in the developed model of the cyber-physical logistics system capable of managing autonomous objects. The developed reinforcement learning models and path planning methods can improve the efficiency and accuracy of route planning for autonomous vehicles, reducing the delivery time and optimizing the use of resources. The developed methods of planning the path taking into account the safety restrictions allow avoiding dangers and accidents on the roads, ensuring safety both for autonomous vehicles and for other road users. Taking into account the constraints when planning the logistics path allows you to increase the efficiency and optimize the delivery processes, reducing the time and resources spent on transporting goods.
The results were implemented in the educational process of the department of information systems and technologies in the preparation of methodological materials for conducting classes on the educational component "Optimal systems".
