Projects

The project aims to explore the adoption of open-source platforms for monitoring production lines and to develop a solution for monitoring Balsnack waffle production line. The objective is to extend the solution to other production lines after successful validation. The project created an integrated production monitoring solution that combines machine vision-based quality control with inputs from sensors. Machine vision facilitates visual quality control and product identification. Product identification will automatically provide the system with preset values that will be compared with inputs from sensors.

Monitoring, recording and analysis of the correct parameters of production equipment at the prototype level using AI.

New methods for producing mat type material from wood springs will be developed based on current equipment and by complementing this with collaborative robot. Possibilities to form the wood spring mat with suitable gripping method will be researched to find an optimal solution to produce this type of packaging material. The found trajectories and kinematics by using the collaborative robot (Omron) will be later used as input for developing next generation of production machine. The generated trajectories and kinematics information from robot programs can be also optimized by using AI tools (Machine Learning) to find more optimized solutions (for trajectory and weaving) considering production speed, travelled trajectories, movement dynamics, created mat density etc. as an input parameters.

The aim of the project is to update the existing subject "Production Digitalization" (6 ECTS), modernize the content of the subject according to the latest rapid developments in the field. New development tools based on virtual (VR) and augmented (AR) reality are introduced on a significant scale, and applications based on them in the areas of processing and production.

A reconfigurable robot system is being developed to check the quality and correct function of machine vision IoT modules. Variations of IoT modules and different are different visual quality control reliability is 80-87%. To make and fix the products, we have designed and manufactured a multi-position special jig, which is controlled by a robot. In order to identify products, a database of IoT modules was created, with the help of which algorithms identify and select the product. Also, quality assessment works based on the same logic. Checked products are divided into two: OK and defective. Defective products are separated and determined to be the actual defect.

The development project of an artificial intelligence-based management system Chemi-Pharm focuses on organizing and visualizing the company's daily production logistics through a smart artificial intelligence-based management system, which contributes to increasing the production throughput and efficiency of production processes.