Application of artificial intelligence in surface irregularities measurements
Grant Agreement No.: PM-II/SP/0104/2024/02
Granting Authority: Ministry of Education and Science
Project duration: 01.02.2024 - 31.01.2026
Project’s Grant Amount: 999 900,00 PLN
Grant Amount for WUT: 216 018,00 PLN
Project Manager at PW: Tomasz Kowaluk
Project Team: Tomasz Kowaluk, Marta Rępalska, Jan Tomasik
Partners: Poznań University of Technology, Cracow University of Technology, Kielce University of Technology
Project description:
The project concerns supporting scientific research conducted by a team comprising researchers from Poznań University of Technology, Warsaw University of Technology, Cracow University of Technology, and Kielce University of Technology, in cooperation with the Central Office of Measures, in the field of applications of artificial intelligence to surface metrology. The funding is intended to support a project aimed at enhancing the research capabilities of the participating institutions and strengthening their intellectual capital in the area of artificial intelligence applications in advanced measurement techniques.
The project aims to continue the work initiated under the first edition of the Polish Metrology programme, focusing on the application of artificial intelligence algorithms, including machine learning, simple neural networks, and convolutional neural networks, to the measurement of surface irregularities. As part of the new initiative, the solutions developed will enable the preliminary and rapid definition of measurement scenarios, including the selection of the measurement method, measurement system, type of measurement probe, initial measurement parameters, as well as the type and degree of data filtering.
The project will make use of a database developed by the research team as part of the first edition of the Polish Metrology Programme. This database will be significantly expanded through the new research activities. The algorithms developed will provide advanced support for metrologists in making non-trivial decisions when designing measurement scenarios, the complexity of which is rapidly increasing in the era of Industry 5.0.
It should be emphasized that the proposed solutions will be fully scalable, both in terms of the input data and the possibility of expanding the scope and granularity of the output data.