INTELLECTUALIZATION OF TRANSPORT SYSTEMS FOR PRINTING SEMI-FINISHED PRODUCTS BASED ON MACHINE VISION AND ARTIFICIAL INTELLIGENCE

Abstract

The article explores the problem of intellectualizing transport systems for printing semi-finished products, which is a crucial stage in the production process of the printing industry. The automation of logistics flow management using machine vision and artificial intelligence significantly enhances the efficiency and accuracy of product transportation. Modern printing enterprises utilize various transport mechanisms, including conveyor transport systems, robotic manipulators, and automated mobile robots (AMRs). One of the key optimization directions is the application of machine vision technologies for monitoring the movement of printing semi-finished products, quality control, and defect analysis. It is noted that machine vision is one of the key technologies that ensure automated control and management in production processes. Its main functions in transport systems for printing semi-finished products include: recognizing the shape and size of products to adjust transportation routes, detecting defects (wrinkles, scratches, irregularities), identifying markings (QR codes, barcodes) for optimal product sorting. The use of high-resolution cameras combined with neural network algorithms enables rapid and accurate real-time image analysis. The system itself includes four interdependent components: a light source, a data acquisition sensor, a processing unit, and a communication module. The article presents a workflow diagram of typical machine vision system components in collaboration with a robotic system for the approval of printing semi-finished products' quality. The proposed system integrates machine learning algorithms, computer vision, and automated actuators, enabling real-time assessment of product conditions and decision-making regarding transport adjustments. The architecture of the intelligent system is described, consisting of a hardware complex (including sensors, video cameras, and controllers), image processing software, and neural network-based data analysis algorithms. Such a system not only recognizes objects but also determines their position, movement speed, and potential deviations from standard parameters. It is determined that artificial intelligence plays a crucial role in improving the efficiency of semi-finished product transportation through the following approaches: machine learning for predicting optimal movement routes, deep neural networks for analyzing large datasets and making real-time decisions, multi-agent systems for coordinating the operation of multiple automated devices. The application of machine vision and artificial intelligence in the printing industry is expected to contribute to the overall digitalization of the sector and enhance its competitiveness. Intelligent transport systems may become a significant step toward the creation of «smart» production facilities, capable of independently adapting to changing operating conditions and ensuring high product quality.