Join us for the next ECE Graduate Seminar, featuring doctoral candidate in electrical engineering Zubaidah Al-Masshadani. She will be presenting "One-Shot Alignment for Transferable Prosthetic Control."
ABSTRACT: Reliable myoelectric control is crucial for practical prosthetic use, yet its performance often degrades with changes in limb position, electrode placement or across different users. This work introduces a one-shot alignment framework based on multi-set canonical correlation analysis, or MCCA, that maintains robust control using only a single trial from a new context. By aligning the latent representations of surface electromyography, or sEMG, signals across sessions, positions and users, the proposed framework eliminates the need for retraining and reduces the required training data. Evaluation under static and dynamic arm conditions, including gesture execution and object interaction, demonstrates robust control across days, positions and users. The proposed minimal-calibration, lightweight model supports reliable prosthetic control despite repositioning, donning/doffing, or user variability, offering substantial potential for rehabilitation and everyday applications.
ZUBAIDAH AL-MASSHADANI is a doctoral candidate in electrical engineering at the University of Central Florida and a Future Faculty Laureate. She is a graduate research assistant in the LIMB and UNARY laboratories at UCF and has served as a teaching assistant in the Department of Electrical and Computer Engineering. She holds a master’s degree in robotics engineering from Florida Polytechnic University, where she studied as a Fulbright scholar, and a bachelor’s degree in mechatronics engineering from the Technical Engineering College Baghdad. Her research focuses on developing generalizable human–machine interfaces for prosthetic control and sensory feedback restoration, leveraging unary computing for data compression and energy-efficient signal processing to advance low-resource, real-time wearable technologies.
Read More