About the Project

This PhD position is focused on the development of edge AI methods for wearable surface electromyography (sEMG) sensing in industrial human-centred applications. The PhD candidate will investigate how real-time AI models can be designed, optimised, and deployed on resource-constrained wearable and embedded platforms for interpreting sEMG signals.

About this opportunity

The research will involve real-time sEMG signal processing, hand gesture recognition, lightweight neural network design, model compression, sensor validation, data collection, and embedded deployment. The candidate will work with wearable sensing systems and adaptive human–machine interfaces, with the aim of developing robust, low-latency, and reliable edge AI solutions suitable for real-world operational environments.

Key responsibilities and duties

• Develop edge AI methods for real-time analysis of wearable sEMG signals.
• Design and optimise lightweight models for hand gesture recognition on embedded and wearable devices.
• Build software pipelines for sEMG preprocessing, denoising, feature extraction, training, and evaluation.
• Apply model optimisation techniques such as quantisation, pruning, distillation, and operator fusion.
• Implement and test real-time inference on edge devices, assessing accuracy, latency, robustness, and energy use.
• Support data collection, experimental validation, analysis, publications, and presentations. Work with academic and industrial collaborators as part of the wider research team.

Funding Notes

The position is fully funded by the Horizon Europe EnduRAI project. The successful candidate will join an international and interdisciplinary research environment, working on cutting-edge AI methods and their application to real-world research challenges. Applications from international candidates are warmly welcome.