Aller au contenuAller au menuAller à la recherche

Séminaire - Aymar Goullet de RugyOptimizing muscle control for neural prosthetics

Abstract :

The Revolutionizing Prosthetics program coordinated by the Defense Advances Research Projects Agency (DARPA) has dramatically boosted the development of actuated upper-limb prostheses, and recent surgical technique that transfers residual nerves to alternative muscle sites has increased the number of muscle signals available to control these devices. What is critically needed now is to develop efficient control strategies linking the two. The goal of my research is to enable this by revealing the key mechanisms by which our hierarchical sensorimotor system normally coordinates muscles. In particular, the aim is to establish how the nervous system learns new patterns of muscle activity, how the original sensorimotor mapping is critical and can be restored, and how amputee could benefit from control schemes integrating subsidiary muscles from the trunk and other limbs.


Selected publications

De Rugy, Aymar, Loeb, Gerald E. and Carroll, Timothy J. (2012) Muscle coordination is habitual rather than optimal. Journal of Neuroscience, 32 21: 7384-7391.

de Rugy, Aymar, Marinovic, Welber and Wallis, Guy (2012) Neural prediction of complex accelerations for object interception. Journal of Neurophysiology, 107 3: 766-771.

Stosic, Jelena, Carroll, Timothy J. Carroll and de Rugy, Aymar (2011) Force synchrony enhances the stability of rhythmic multi-joint arm coordination. Experimental Brain Research, 213 1: 117-124.

Woolley, DG, de Rugy, A, Carson, RG and Riek, S (2011) Visual target separation determines the extent of generalisation between opposing visuomotor rotations. Experimental Brain Research, 212 2: 213-224.

Scientific focus :

The primary aim of Dr. de Rugy's research is to understand the control of rhythmic coordination in Human, through a dialogue between experiment and modeling. The general architecture of a computational model composed of neural oscillators coupled to mechanical-effector systems provides the framework for experimental testing. As such, the modeling work serves both as an implementation of mechanisms revealed by experiments and as a basis for the formulation of novel working hypotheses.

So far, the formal model accounts for the behavior observed in the context of hand-held pendulums oscillated individually by participants, and when two individuals attempt to synchronize with each other (i.e., inter-personal coordination). This theoretical and experimental work seeks to unravel the influence of the mechanical properties of the effector systems involved, and generate questions/hypotheses concerning the nature of the neural systems that are engaged, particularly with respect to the utilization of proprioceptive feedback. Ongoing projects are being conducted in the context of bimanual and multijoint rhythmic coordination.

A parallel program of research focuses upon the interplay between the structure of the neuromuscular-skeletal system and intersegmental dynamics in rhythmic multijoint coordination. A robot arm is utilised to provide online control and manipulation of intersegmental dynamics during rhythmic multijoint arm movements. This manipulation is designed to identify our capability to exploit intersegmental dynamics when using our limbs to interact with the environment, and to isolate the related constraints imposed by the structure of the neuromuscular-skeletal system. These constraints include the complex configuration of muscles (some of them being bi-functional) and the associated reflex pathways. Neurophysiological techniques (e.g., fine wire electromyography, transcranial magnetic stimulation) are being used in combination with biomechanical analyses to identify the specific neuromuscular-skeletal constraints that are involved, and their influence on the observed behavior.

Aymar de Rugy joined the School in January 2004 as a University of Queensland postdoctoral research fellow. Prior to that, he completed a two-year research fellowship at the Pennsylvania State University (USA). Aymar completed a PhD at the University of the Mediterranean ( France) for which the topic of research was the control of visually guided locomotion.

Jean René Cazalets (jean-rene.cazalets @ u-bordeaux2.fr)