Regulation of FES-induced grasp force based on cutaneous nerve signals: experiments and modelling

We investigated the control of electrically induced hand grasp by natural sensors located in the skin of the index finger. A tetraplegic volunteer was implanted with an eight-channel muscle stimulator, providing hand grasp, and a nerve signal recording cuff electrode placed on a branch of the palmar digital nerve deriving from the median nerve and innervating the radial aspect of the index finger. The recorded nerve signal contained information that could be used to detect slips and further to increase the stimulation intensity to stop a slip. A mathematical model of the experiments was employed to understand the behaviour of the hand grasp system, to extrapolate data obtained in experiments, to test parameter influence on the length of a slip, and to refine the control algorithm. The model considered the mechanical conditions causing a slip, the major properties of the musculo-skeletal system of the hand in lateral grasp, and the relation between the processed nerve signal and the slip velocity. The objective of this study was to investigate the control algorithm in both experiment and simulation. We tested reactions on a slip with different fixed increases in stimulator command and an algorithm with adaptive increase in stimulator command based on the amplitude of the processed nerve signal. Further, the application of a number of maximum stimulator commands with double the instantaneous stimulation frequency as an initial reaction to a slip was investigated. We tested three different surfaces of the held object, representing different frictional conditions of the skin-object contact. A discussion of the slip-based controller concludes the paper.