Please check our positions below. If you are interested in joining our lab, contact nsquared@fau.de with a short description of your scientific background and what you would like to do, e.g. bachelor/master thesis.
Student Assistant for Teaching Activities
We are searching for a student assistant (7h/week) to support in the organisation of our lectures and practical exercises, as well as in the organisation of our laboratory. Your tasks will include the creation/updating of lecture/exercise slides, supporting students during practical exercises, evaluating and grading exercise submissions and other tasks related to our teaching activities. If there is no current task to do related to teaching, you will help in the organisation of our research group, as well as assist in our various research projects. Depending on your skills and background, this can include mechanical design, electronic design, assistance in research studies and data acquisitions, programming and data analysis, etc.
Start date: 01.04.2026
Tasks:
- Teaching-related tasks (see above)
- Organisational tasks
- Assistance in research projects/data acquisitions
Requirements:
- Bachelor (3rd semester and above) or Master student (1st/2nd semester)
- Proficiency in both written and spoken English and German (B2+)
- High proficiency and experience in the use of MS Office software
- Ideally: Experience in Biomedical Signal processing in MATLAB or Python
- Ideally: Experience in Mechanical/Electronical Engineering
- Interest in committing to at least a 1-year position with the option of extention
Supervisor:
Marius Oßwald, M. Sc.
Application:
Please provide a short CV and transcript of records to nsquared@fau.de.
Integration of online model recalibration into cursor control application
We are offering a thesis project focused on the integration of retraining across sessions as well as online model recalibration during real-time cursor control based on sEMG data recorded from legs and forearm.
Tasks:
- Review relevant literature
- Explore performance degradation and ways for model recalibration
- Integrate validated findings into existing applications
- Conduct experiments and evaluate performance
Requirements:
- Interest in biosignal processing and neural interfacing
- Very good programming experience in Python (experience with PySide6, SciPy, PyTorch/Scikit-learn is a plus)
- Solid knowledge of classical machine learning and deep learning
- Ability to work independently and systematically
Supervisor:
Amin Olamazadeh, M.Sc.; Annika Ritter, M.Sc.
Application:
Please provide a short CV and transcript of records to nsquared@fau.de.
Validation Study of a custom-build 3D Hand Dynamometer
Planning, designing and conducting a validation study of a custom-built 3d hand dynamometer. The device is designed to measure isometric finger forces in various directions. The validation study will assess the reliability of the dynamometer across multiple recording sessions for all five digits during both single and multi-digit contractions.
Tasks:
- Develop a comprehensive plan for the validation study
- Design and implement experimental protocols to evaluate the dynamometer’s performance
- Collect and analyse data to assess the reliability and accuracy of the dynamometer
- Document the study design, methods, results, and conclusions
Requirements:
- Solid python skills
- Solid knowledge in CAD is beneficial
- Experience with git is beneficial
- Ability to work independently and sound time management skills
Supervisor:
Charlotte Pradel, M.Sc.
Application:
Please provide a short CV and transcript of records to nsquared@fau.de.
Development of an AI-Based Multimodal Control System for Intelligent Wheelchairs Using EMG Signals
This master’s thesis is jointly supervised by the Chair of Manufacturing Automation and Production Systems (FAPS) and the n-squared lab.
Motivation:
Mobility impairments have far-reaching effects on the daily lives of affected individuals and often lead not only to physical limitations but also to psychological stress. For many people, a wheelchair represents the only way to participate in everyday life. However, classical control concepts reach their limits in the case of complex medical conditions (e.g. high-level spinal cord injuries), since the remaining muscle activity is often insufficient to operate a joystick precisely. The EMGRoll project addresses this problem through a sensor kit that uses electromyography (EMG) for control. While traditional approaches are often based on rigid rule-based systems, modern robotics offers new possibilities through methods such as Reinforcement Learning (RL) and Imitation Learning. The goal is no longer to manually program rules for every situation, but instead to train a multimodal AI model. This model should be able to directly fuse the noisy and highly individual EMG signals of the user with environmental data (e.g. LiDAR/camera). The result is an adaptive “shared autonomy” system that understands the user’s intention and intelligently translates it into safe driving commands
Objectives:
The objective of this thesis is the design, implementation, and evaluation of a deep-learning-based navigation approach that processes EMG signals and sensor data within a single neural network. The thesis includes the following key components:
- Literature Review: Evaluation of the state of the art in Deep Reinforcement Learning (DRL) and Learning from Demonstration (LfD) for assistive robotics.
- Conceptual Design: Development of a multimodal network architecture that processes physiological signals (EMG) and exteroceptive sensor data (LiDAR/camera) as inputs.
- Implementation: Development of the training environment (e.g. simulations such as Gazebo, Unity, or Isaac Sim) and implementation of the agent in ROS2.
- Training & Evaluation: Training of the model using RL or imitation learning and validation of its performance in comparison to classical approaches.
- Transfer: (Optional / depending on scope) Sim-to-real transfer to a real wheelchair demonstrator at FAPS.
Further information is available upon request.
Applications should be sent by email, including a current transcript of records and a CV.
References:
- Reinforcement Learning Based User-Specific Shared Control Navigation in Crowds
- Learning from demonstration for locally assistive mobility aids
- Learning Shared Control by Demonstration for Personalized Wheelchair Assistance
- Shared control of a brain-actuated intelligent wheelchair
- Shared control methodology based on head positioning and vector fields for people with quadriplegia
Supervisor:
Matthias Kalenberg, M.Sc.
Amin Olamazadeh, M.Sc.
Application:
Please provide a short CV and transcript of records to matthias.kalenberg@faps.fau.de or amin.olamazadeh@fau.de.
Validation Study of a custom-build 3D Hand Dynamometer
Planning, designing and conducting a validation study of a custom-built 3d hand dynamometer. The device is designed to measure isometric finger forces in various directions. The validation study will assess the reliability of the dynamometer across multiple recording sessions for all five digits during both single and multi-digit contractions.
Tasks:
- Develop a comprehensive plan for the validation study
- Design and implement experimental protocols to evaluate the dynamometer’s performance
- Collect and analyse data to assess the reliability and accuracy of the dynamometer
- Document the study design, methods, results, and conclusions
Requirements:
- Solid python skills
- Solid knowledge in CAD is beneficial
- Experience with git is beneficial
- Ability to work independently and sound time management skills
Supervisor:
Charlotte Pradel, M.Sc.
Application:
Please provide a short CV and transcript of records to nsquared@fau.de.
Validation Study of a custom-build 3D Hand Dynamometer
Planning, designing and conducting a validation study of a custom-built 3d hand dynamometer. The device is designed to measure isometric finger forces in various directions. The validation study will assess the reliability of the dynamometer across multiple recording sessions for all five digits during both single and multi-digit contractions.
Tasks:
- Develop a comprehensive plan for the validation study
- Design and implement experimental protocols to evaluate the dynamometer’s performance
- Collect and analyse data to assess the reliability and accuracy of the dynamometer
- Document the study design, methods, results, and conclusions
Requirements:
- Solid python skills
- Solid knowledge in CAD is beneficial
- Experience with git is beneficial
- Ability to work independently and sound time management skills
Supervisor:
Charlotte Pradel, M.Sc.
Application:
Please provide a short CV and transcript of records to nsquared@fau.de.
Python GUI Toolbox for Biosignal Data Analysis
We are looking for a Medical Engineering student for a 10 ECTS project to develop a Python-based GUI toolbox for EMG data analysis.
The project involves developing a Python (PySide6) GUI toolbox for interactive biosignal data analysis. The tool will support GUI-controlled signal processing, feature extraction, flexible plotting, label visualization, and exploratory analysis, enabling users to experiment with different methods and parameters.
Requirements:
- Strong Python programming skills
- Experience with Python GUI development (PySide6 / PyQt)
- Familiarity with scientific Python tools (NumPy, SciPy, Matplotlib)
- Interest in biosignal processing and data analysis
- Git experience is beneficial
- Ability to work independently
Supervisor:
Annika Ritter, M.Sc.
Amin Olamazadeh, M.Sc.
Application:
Please provide a short CV and transcript of records to nsquared@fau.de.