Bachelor-/ Master Thesis: »Tool Wear meets Federated Learning«

Rheinisch-Westfälische Technische Hochschule Aachen; Aachen; Vor Ort; EUR 40.000 - 60.000; Vollzeit

Zusammenfassung

A leading research institute in Aachen is seeking a student for a thesis project on federated learning for tool wear detection. Candidates should be studying production or mechanical engineering, and possess programming skills in Python along with machine learning experience. This role offers hands-on experience using advanced technologies in a collaborative environment.

Leistungen

Flexible working hours

Professional supervision

Access to state-of-the-art facilities

Qualifikationen

• Studying production engineering, mechanical engineering, or a comparable subject.
• Initial programming (Python) and Machine Learning (PyTorch, Scikit-learn) experience is required.
• A high degree of initiative, independence, and motivation.
• Good language skills in German or English.

Aufgaben

• Analyze acquired microscopic and video tool wear data.
• Investigate image processing techniques such as spatial filtering, morphological techniques, and segmentation.
• Hands-on development of pre-processing, AI-training, and evaluation pipelines.
• Evaluate centralized, individual, and federated learning scenarios for tool wear detection.

Kenntnisse

Programming (Python)

Machine Learning (PyTorch, Scikit-learn)

Image Processing Techniques

Good language skills in German or English

Ausbildung

Production engineering or mechanical engineering

Jobbeschreibung

The Fraunhofer-Gesellschaft (www.fraunhofer.com) currently operates 76 institutes and research institutions throughout Germany and is the world’s leading applied research organization. Around 32,000 employees work with an annual research budget of 3.6 billion euros.

At the Fraunhofer IPT in Aachen, we work with more than 530 employees every day to make the production of the future more digital, more flexible and more sustainable. In the department "High-Performance cutting", we deal with high-quality requirements in the metal cutting industry, especially in highly regulated sectors such as aerospace.

As part of your thesis, you will investigate federated learning for decentralized AI model training for tool wear detection and measurement in milling processes within the FL4AI project. A custom dataset has been acquired, consisting of microscopic tool wear images and CNC-integrated camera videos. Your task will be to generate AI pipelines to detect and measure flank wear according to ISO DIN 8688-2 in centralized, individual, and federated learning scenarios. Here you partly work on your tasks on-site in our institute/ machine park.

What you will do

• Analyze acquired microscopic and video tool wear data
• Investigate image processing techniques such as spatial filtering, morphological techniques, and segmentation
• Hands‑on development of pre-processing, AI-training and local and global evaluation pipelines
• Evaluate centralized, individual, and federated learning scenarios for tool wear detection and measurement

What you bring to the table

• You are studying production engineering, mechanical engineering, or a comparable subject
• Initial programming (python) and Machine Learning (pytorch, and scikit-learn) experience is required
• A high degree of initiative, independence, and motivation
• Good language skills in German or English

What you can excpect

• Ideal conditions for practical experience alongside your studies
• Professional supervision and collaboration in a dedicated team
• A state‑of‑the‑art machine park equipped with edge cloud systems and 5G infrastructure
• Flexible working to combine study and job in the best possible way

Interested? Apply online now. We look forward to getting to know you!

For any further information on this position please contact:
Gustavo Laydner de Melo Rosa Eng. Mec.
Research assistant »High Performance Cutting«
Phone: [phone]