Applied AI

Industrial applications of artificial intelligence across manufacturing, quality control, and transport domains.

Research Overview

This research theme encompasses the application of AI and machine learning methods to real-world industrial problems across diverse domains, including manufacturing, quality control, and transport systems.

Key application areas include:

• Manufacturing process optimization: Using invertible neural networks for machine parameter prediction in heat-setting processes and fused deposition modeling.
• Quality monitoring: Applying Generative Adversarial Networks for non-destructive quality assessment in large-scale casting.
• Neural network-based process modeling: Leveraging deep learning for high-dimensional data scenarios in industrial production.
• Reinforcement learning: Developing simulation-as-a-service platforms and transfer learning approaches for robotic manipulation tasks.

Related Publications

• Objectifying Machine Setup and Parameter Selection Using Invertible Neural Networks (2023) Müller, K., Posada-Moreno, A., Pelzer, L., Gries, T. — FAIM 2022 (Springer LNME) Details | DOI

• Process Parameter Prediction for Fused Deposition Modeling Using Invertible Neural Networks (2023) Pelzer, L., Posada-Moreno, A.F., Müller, K., Greb, C., Hopmann, C. — Polymers Details | DOI

• Qualitätsüberwachung mit GANs als Alternative zur Angussprobe für den Großguss (2023) Weber, F., Yao, Y., Horbach, L., Posada Moreno, A.F., Bezold, A., Broeckmann, C. — Tagung Werkstoffprüfung Details | DOI

• Neural-Network-Based Classification of Commercially Available Fish Fillets (2021) Bragantini, G., Posada-Moreno, A.F., et al. — Proceedings of Machine Learning Research Details

• Simulation-as-a-Service for Reinforcement Learning Applications (2020) Scheiderer, C. et al. — Procedia Manufacturing Details | DOI

• Transfer of Hierarchical Reinforcement Learning Structures for Robotic Manipulation Tasks (2020) Scheiderer, C., Mosbach, M., Posada-Moreno, A.F., Meisen, T. — CSCI Details | DOI