I am a PhD student at the Control- and Cyber-Physical Systems Lab, TU Darmstadt. My research is focuses on using deep learning for transition detection for exosuit using multimodal data fusion of different sensor modalities. I am particularly interested in what type of sensory data you need to predict user intentions and how to design deep learning architectures that can learn meaningful representations from this data. I am also interested in the broader question of how we can use machine learning to better understand body and brain functions and develop new brain-computer interfaces and human-in-the-loop systems.
My interests include EEG/IMU/EMG analysis, representation learning, robotics, signal decoding, medical AI, human-in-the-loop and brain-computer interfaces. I enjoy building research software and applying modern machine learning methods to challenging biomedical problems.
My current work investigates how transition in movement can be detected from multimodal sensor data for exosuit control. With the intent of developing more intuitive and responsive assistive devices.
Research on multimodal sensor data fusion for exosuit control.
Working Title: "From Signal to Action: Learning-Based Fusion of High Dimensional Sensory Data for Robotic Control".
Development of graph-based electrical grid modeling tools using Python, SQL, Pandas, and NetworkX.
Development of a desktop application for low-voltage grid planning using React, Electron, Python, and MATLAB.
During my studies and research career, I have worked as a teaching assistant for several programming courses, supporting students with software development, algorithms, and practical coding assignments.