Students Projects

Master Thesis: Analysis of Neural Input Dimensionality in Human Motor Control

The control of movement depends on the coordinated activity of many motor neurons. Recent advances in signal analysis have made it possible to identify patterns of shared activity between motor units, providing insights into how the nervous system organizes muscle control. In this project, you will develop and evaluate a new signal processing methodology to analyze the input received by motor units. The focus will be on building computational tools that can process high-resolution data.  [Contact us]

This work has applications in neuroscience, rehabilitation, and human–machine interfacing.

Your Qualifications

• Strong programming skills (e.g., MATLAB, Python, or similar).
• Strong background in signal processing and machine learning/deep learning methods.
• Interest in human motor control and neural signal analysis.
• Independent working style and motivation to solve interdisciplinary research questions.

Relevant Literature

• Del Vecchio, Alessandro, et al. "The forces generated by agonist muscles during isometric contractions arise from motor unit synergies." Journal of Neuroscience 43.16 (2023): 2860-2873.
• Hug, François, et al. "Correlation networks of spinal motor neurons that innervate lower limb muscles during a multi‐joint isometric task." The Journal of physiology 601.15 (2023): 3201-3219.
• Hug, François, et al. "Common synaptic input, synergies and size principle: Control of spinal motor neurons for movement generation." The Journal of physiology 601.1 (2023): 11-20.

Affordance-Aware Multi-Fingered Manipulation via LLM

We are seeking a highly motivated student to join our team. In this Master's thesis, you will investigate how affordance-aware representations derived from LLMs can guide multi-fingered hands through object handling tasks.  [Contact us]
Keywords: LLM, Robotic Manipulation, Reasoning, Robot learning

Your Tasks
   • Review state-of-the-art in LLM-based robotic manipulation
   • Implement and evaluate semantic grasp reasoning modules
   • Design and run experiments on a robotic hand platform, with an existing dexterous hand
   • Evaluation of the capabilities and limits of such tools

Your Qualifications
   • Enrolled in a Master’s program (STEM or related fields)
   • Passionate about machine learning and robotics
   • excellent in python (experience with PyTorch or JAX is a bonus)
   • Self-motivated working

Relevant publication:
Mirjalili, Reihaneh, et al. "Lan-grasp: Using large language models for semantic object grasping." arXiv preprint arXiv:2310.05239 (2023)

Master Thesis: Decoding Movement Intention from Motor Unit Activity

In this Master’s thesis, you will develop a biologically informed approach for decoding movement intentions from muscle activity. Using high-density surface electromyography (HD-sEMG) to capture activity from individual motor units, you will work on translating this neural information into real-time control signals for gesture recognition. This project bridges neuroscience and human-machine interfacing, aiming to create more natural and physiologically grounded strategies for controlling neuroprosthetic devices and myoelectric interfaces.  

[Contact us]

Your Tasks

• Review the state of the art in motor unit decomposition, and motor unit based real-time control
• Decode MU activity in real time using a pre-existing framework
• Integrate motor unit features into classification/regression pipelines for movement intention decoding
• Design and run real-time biofeedback experiments to evaluate decoding accuracy, robustness, and user adaptation

Your Qualifications

• Proficiency in Python and Matlab
• Background in signal processing, machine learning, and neural data analysis
• Preferably, knowledge in the processing of sEMG and motor unit data
• Strong interest in neural interfaces, motor control, and human-machine interaction

Relevant Literature

• Chen, Chen, et al. "Real-time hand gesture recognition by decoding motor unit discharges across multiple motor tasks from surface electromyography." IEEE Transactions on Biomedical Engineering 70.7 (2023): 2058-2068.
• Chen, Chen, et al. "Simultaneous and proportional control of wrist and hand movements by decoding motor unit discharges in real time." Journal of Neural Engineering 18.5 (2021): 056010.

Exploring neural content and spatial-temporal information from high density surface electromyographic sensors

This master thesis focuses on extracting spatial-temporal features from high-density surface electromyographic (HD-sEMG) signals to decode motor intention, enabling more precise and dexterous human-machine interfacing. The work involves analyzing signal features to achieve accurate classification/regression of different movements, which can then be applied to control bionic limbs. Studies investigating motor modularity in humans and motoneuron activity are also highly encouraged, as they can provide insights into optimizing control strategies for assistive devices. [Contact us]

Your Tasks

• Review the state of the art in high-density EMG signal processing and neural decoding
• Analyze spatio-temporal signal features from HD-sEMG open datasets
• Implement and propose new methods to map neural information to movement intent
• Investigate modularity in motor control and adaptation

Your Qualifications

• Enrolled in a Master’s program (STEM, computer science, neuroscience, biomedical engineering, or related field)
• Background in signal processing and/or machine learning

Open applications

We welcome highly motivated students who wish to contribute but find that their interests are not reflected in our current projects. 

If you have an idea in mind, please reach out to us with your proposal. Always include your CV and academic transcripts to allow us to evaluate your application. [Contact us]