Bachelor / Master Thesis: Development of a Multi-Agent System for the interaction with System Models

Webseite Institute of Machine Elements and System Engineering

The Institute of Machine Elements and System Development researches the fundamental structural and tribological behavior of machine elements and depicts this in experimentally validated model descriptions. These model descriptions are used to analyze and design the functional, loss and noise behavior of entire technical systems with a focus on drive technology. The developed models are also used to research and develop methods of model-based systems engineering as a central element of future, inductive product development processes.

This work explores the transformative potential of AI in product development through the use of agents for natural language interaction with system models. Based on the existing MBSE model, state-of-the-art AI techniques are implemented to enable seamless access to complex system information and make technical data more intuitive and usable. This work provides an opportunity to redefine the way engineers and stakeholders interact with system models.

In the KIMBA research project, we are working with leading OEMs and Tier1 suppliers in the automotive industry to develop innovative approaches to the digitalization of product development. The focus is on the modeling and system-wide linking of information to create continuous data flows.

Tasks:

  • Concept and evaluation framework development
  • Exploration of different AI architectures
  • Design and implementation of an (Multi-) Agent System to interact with system models via natural language input
  • Evaluation of the results
  • Written documentation and presentation in the form of a final presentation and a demonstrator

Prerequisite:

  • Independent, reliable way of working
  • Good programming skills, preferably in Python
  • Experience with AI/ML frameworks (e.g. TensorFlow, PyTorch, Hugging Face, LangChain etc.) an advantage
  • Depending on the focus, knowledge of SQL databases
  • Willingness to acquire a basic understanding of MBSE and SysMLv2

We offer:

  • Flexible organization of work priorities
  • Rapid processing options
  • Intensive support
  • Immediate start or by arrangement
  • Very good working atmosphere

 

Auf deine aussagekräftige Bewerbung per E-Mail freut sich:

Vincent Quast, M. Sc. RWTH
Institut für Maschinenelemente und Systementwicklung

Eilfschornsteinstraße 18, 52062 Aachen
vincent.quast@imse.rwth-aachen.de

Um sich für diesen Job zu bewerben, sende deine Unterlagen per E-Mail an vincent.quast@imse.rwth-aachen.de