Login
Help
Documentation
Back to
News Board
Browse in News
Siemens AG Master's Thesis: Safe Reinforcement Learning for Real-World Control Systems
15.10.2025, Diplomarbeiten, Bachelor- und Masterarbeiten
This project focuses on implementing safe reinforcement learning algorithms for real hardware control systems. Students will work on bridging the simulation-to-reality gap by deploying AI control methods on edge computing platforms integrated with PLC systems, while ensuring safety and stability.
Location: Siemens Technology Center, Garching
Start Date: Flexible between January and March
This thesis project addresses challenges in deploying AI-control systems on real hardware systems, specifically in the safe adaptation of models to further improve performance.
The project focuses on implementing and validating safe reinforcement learning methods for control that can bridge the sim-to-real gap while maintaining safety. You will work with state-of-the-art algorithms that use control-theoretic principles to ensure system stability and enable safe exploration in real-world environments.
Research Objectives
- Algorithm Implementation: Implement published algorithms for continuous control systems with theoretical safety guarantees
- Hardware Integration: Deploy learning algorithms on edge computing platforms integrated with PLC control systems
- Performance Evaluation: Compare learning approaches on real hardware
Requirements
- Master's student in Electrical Engineering, Computer Science, Robotics, Mechatronics, or related field
- Strong foundation in control systems, stability theory, and dynamical systems
- Strong programming skills in Python, ideally with machine learning libraries
- Familiarity with Linux systems and embedded programming
- Familiarity with reinforcement learning concepts
Preferred Qualifications
- Previous work with PLCs, industrial networks, or automation systems
- Experience with industrial communication protocols
What We Offer
- Work in an interdisciplinary research group of technical experts focused on Design, Control & System Integration in the field of Power Electronics and Automation
- Apply cutting-edge digital technologies, in the fields of artificial intelligence, machine learning, software development, simulation and automation to solve real-world industry problems.
- Flexibility in defining the project scope based on your interests.
How to Apply:
Send an email to olivia.garland@siemens.com with the following:
- Your CV
- A brief introduction outlining your background and motivation
- Academic Transcripts
Kontakt: olivia.garland@siemens.com
Print
RSS 
