TUM Logo

Anomaly Detection in Additive Manufacturing

Anomaly Detection in Additive Manufacturing

Supervisor(s): Nikolai Puch
Status: open
Topic: Anomaly Detection
Type of Thesis: Masterthesis
Thesis topic in co-operation with the Fraunhofer Institute for Applied and Integrated Security AISEC, Garching

Description

Master Thesis

Anomaly Detection in Additive Manufacturing

Motivation and Task

The increasing use of additive manufacturing (AM) technologies, colloquially known as 3D printing, in an Industry 4.0 setting has made it a topic of interest in IT security and forensics. Especially the use of AM for safety-critical parts, for example in the SpaceX SuperDraco engine or parts of an Airbus jet engine, underlines the importance of such research. Compared to the more process-invasive method of directly measuring all parts, side channels can be monitored to detect attacks on an AM production line without major changes in the process. Byeffects like vibration, power consumption, electromagnetic, or thermal emissions can be used for this purpose. This information can be evaluated using techniques such as machine learning (ML) algorithms to detect deviations from the regular production flow.

In this thesis an overview of different methods for detecting anomalies in AM should be compiled. Based on this summary, the side channels of the AM process should be evaluated for their usefulness in detecting attacks on AM via ML. Finally, a proof of concept should be developed for the best suited side channel.

Due to the nature of the task, presence at the institute in Garching by Munich is required. We ask you take this fact into account with your application.

Prerequisites

The following list of prerequisites is neither complete nor binding, but shall give you an idea, what the topic is about:

• Interest in AM (i.e. 3D Printing) and security in manufacturing
• Good general programming skills (ideally Python)
• Ideally experience with ML or an interest in familiarising oneself with it

• Ability to work self-directed and systematically

Please attach a current grade sheet and a short CV to your application so that we can assess your qualification for the topic of your choice.

The thesis can be written in English or German.

Contact

Nikolai Puch

Telefon: +4989322-9986-142
E-Mail:
nikolai.puch@aisec.fraunhofer.de
Fraunhofer Research Institution for Applied and Integrated Security (AISEC) Department Product Protection & Industrial Security
Lichtenbergstraße 11, 85748 Garching (near Munich), Germany
https://www.aisec.fraunhofer.de