Results

The "ICNAP Study Report 2023" presents the results of five collaborative research and development studies conducted by the International Center for Networked Adaptive Production (ICNAP) and its industry partners in 2023. They condense current findings, facts, figures, and results from discussion, workshops, research, and development collaborations between researchers and partner companies in the ICNAP community.

The project involved collecting a set of use cases from scientific publications and ongoing scientific and industrial projects, wherein a two-dimensional clustering was performed based on compute objectives and sectors of research.

This project explored a non-invasive approach to individual part recognition in industrial production using camera-based tools and locality-sensitive hashing (LSH) algorithms.

In this project, a computer vision-based system for crop quality control has been developed, facilitating continuous and non-destructive preharvest quality assessment for efficient and sustainable indoor production of plant-based foods.

The Evolve project aimed to develop a software application utilizing Bayesian Optimization for custom process optimization in two distinct cases: ultrashort pulse laser structuring of metallic surfaces and the cultivation of plant suspension cells in a batch fermentation bioreactor.

This study identifies opportunites for the use of AI in optimizing process designs.

For many processes, companies require a secure transfer across company boundaries. In this project, we built a permissioned blockchain network that enables a secure transfer between two different companies.

In this project, active and semi-supervised learning principles were used to test efficient and interactive training of deep learning systems.

This project researched and evaluated approaches to image data synthesis for machine learning systems in manufacturing, which were implemented for two use cases: tool classification and classification of surface structures after laser polishing.

How can energy consumption be monitored and mapped for energy modelling?