Restoring movement through neuroprostheses : a robotics problem?

David Guiraud obtained his engineering degree from the Ecole Centrale de Paris and his master's degree in molecular biological analysis in 1990; his doctorate in science focused on artificial neural network-based control for exoskeletons and neuroprostheses used to treat spinal cord injuries. An Associate Professor of Applied Physics at Montpellier IUT until 2001, he then founded the DEMAR team at LIRMM with Inria Sophia Antipolis Méditerranée in 2004. During this period, he was involved in a number of European projects piloted by the Montpellier Faculty of Medicine, aimed at restoring lost motor functions through the use of implanted stimulation. Awarded the CNRS Bronze Medal in 2005 and the Grand Prix de l'Académie des Sciences in 2010, his multi-disciplinary work has been rewarded and recognized in the field of biomedical engineering. For over 15 years, David Guiraud has been working on models of the deficient sensory-motor system and studying palliative technological solutions based on neuroprostheses, particularly implantable ones. With particular regard to the restoration of motor functions, the theoretical tools and approaches used in automatic control are similar to those used in robotics, so that his work is naturally linked to that of health robotics. David Guiraud attaches particular importance to results having a real clinical impact, and a large part of his work is closely linked to the medical world. Restoring movement through neuroprostheses: a robotics problem? The simulation, synthesis and control of movement, and the processing of signals, particularly electrophysiological signals, underpin developments and research in this field, particularly when it comes to evaluating and quantifying the contribution of neuroprostheses. These approaches are also essential to the implementation and adjustment of these particularly complex systems, where the actuator - the muscle - the sensors - natural or artificial - and the control - sometimes combining voluntary action and artificial control - form a whole reminiscent of all the elements of a robot. However, each element is profoundly different in nature and far more complex in behavior, so that a "simple" transposition is not possible. Based on a review of the current state of the art, this presentation will attempt to identify possible synergies and inevitable points of divergence between the two fields, particularly in the case of restoring movement to spinal cord injured patients.