The AI4QAM-project aims to create user-friendly AI solutions for automated quality control in the manufacturing industry.

AI technology improves automation in industry

State-of-the-art AI technology will improve automation in the manufacturing industry. That is the goal of a research project led by RISE and the School of Engineering (JTH) at Jönköping University. Vinnova supports the project with SEK 8.47 million, of which SEK 3.9 million goes to JTH.

The research project is called “(End-to-End) AI for Quality Assurance in Manufacturing” (AI4QAM) and aims to create user-friendly *End-to-End AI solutions for automated quality control in the manufacturing industry. To achieve good results, different AI techniques will be used, which also reduces environmental impact and resource consumption.

Improve quality assurance processes

The overall leadership for the project lies with RISE, but JTH will be responsible for some of the work packages within AI and robotics at the partner companies in the project.

“The main goal is to help companies with their AI transformation moving towards Industry 5.0 - mainly by focusing on implementing state-of-the-art AI technology to improve automation in manufacturing. For Swedish industry to remain competitive in an already strained manufacturing market, this is a necessary step to keep level with the rapidly developing technology that will change industries worldwide. AI technology has the potential to speed up production, automate complex steps, reduce resource consumption, and improve waste management in industry, as a complement to human operators,” says Milad Ashour Pour, Assistant Professor doing his research in automation and robotics at JTH and work package manager for the AI4QAM project.

Shows JTH's strengths

The sum of SEK 3.9 million to JTH will be distributed over three years and primarily used for personnel costs within the project.

Kerstin Johansen, Associate Dean of Research at JTH, thinks that the End-to-End project shows how JTH can combine its strengths and competence to the great advantage of its partners. Namely, the project covers three departments at JTH and two areas of strength within the school's research and education environment SPARK. In addition, co-operation from Husqvarna is included as well as collaboration with RISE and three of JTH's strategic partners that are included as partner companies.

Facts

The following researchers are involved in the project from JTH

Milad Ashour Pour, Assistant Professor in automation and production technology.
Ghazaleh Azizpour, Visiting Assistant Professor at JTH and Technology Innovation Lead - Automation & Robotics at Husqvarna AB.
Maria Hedblom, Assistant Professor, Computer Science.
Maria Riveiro, Professor, Computer Science.

The following partner companies are involved in the project:

School of Engineering
RISE (Research Institutes of Sweden AB)
Husqvarna AB
Scania AB
Thule Group AB
Enodo Robotics AB
PVI Hydroforming AB

*End-to-End AI is a technique where the model learns all the steps between the initial input phase and the final result. This is an in-depth learning process where all the different parts are trained simultaneously instead of one after the other.