Max Robotics — Global Service Robot Directory + US Certification

About

Marcello Zanghieri is a researcher specializing in human-machine interfaces, embedded machine learning, and bioelectrical signal processing, with a particular focus on surface electromyography (sEMG)-based hand gesture recognition and control. His work sits at the intersection of neural networks, edge computing, and rehabilitation technology, addressing the practical challenges of deploying intelligent systems on resource-constrained hardware. Zanghieri's most influential contribution, "sEMG-based Regression of Hand Kinematics with Temporal Convolutional Networks on a Low-Power Edge Microcontroller" (2021, 29 citations), demonstrated that sophisticated deep learning models could be effectively compressed and executed on ultra-low-power microcontrollers — a critical step toward viable wearable prosthetic devices. Complementing this, his work on on-device incremental learning (15 citations) tackled the persistent problem of sEMG signal variability over time, enabling gesture recognition systems to adapt dynamically without cloud dependency. Further research explored neuromorphic-inspired spike reconstruction and end-to-end deep learning pipelines for robust gesture classification. Collectively, Zanghieri's contributions advance the feasibility of intelligent, real-time prosthetic and rehabilitative interfaces, bridging the gap between state-of-the-art machine learning and the stringent constraints of real-world wearable deployment.