Context Aware Control: integrating Machine Learning and physics Informed modelling in Soft Wearable Robotics
Lorenzo Masia

back to Overview

Date: Wednesday, 26.06.2024 15:20-17:00 CET

Location: Building S1|15 Room 133

In case you are interested in a 1-on-1 meeting or meal with the speaker, please contact the coordinator Tobias Thomas:

Abstract:

In the evolving landscape of assistive technology, soft wearable exosuits have emerged as a pivotal innovation, distinguishing themselves from their rigid counterparts, the exoskeletons. Yet, mastering soft structures is not an easy task: seamlessly coordinating robotic assistance and human motion requires to compensate not only for the non-linear behaviours of the device but also correctly interpreting the physiological signals which take part to the main control loop.

My presentation will be focused on the latest progress from my group over the last five years in Heidelberg University: I will describe all the approaches we have in-house developed to obtain compact, robust, reliable and efficient exosuits. I will digress on the importance of embedding biomechanical modelling in the control strategies in order to tailor the interaction of the machine with the wearer´s biomechanics, targeting human performance augmentation in physical tasks, like lifting or improvement of running endurance.

Emphasis will be dedicated in illustrating a completely novel approach named “Context Aware Control”, which merges classic control strategies and machine learning, including artificial vision, to optimize modulation of the assistance. Such a solution provides our exosuits with the unique capacity to understand and respond to different external contexts or environmental changes, dramatically enhancing user ‘symbiosis with the wearable robotic systems.