Characterizing the human-robot haptic dyad in robot therapy of stroke survivors

Purpose – The working hypothesis, on which this paper is built, is that it is advantageous to look at protocols of robot rehabilitation in the general context of human-robot interaction in haptic dyads. The purpose of this paper is to propose a new method to detect and evaluate an index of active participation (AC index), underlying the performance of robot-assisted movements. This is important for avoiding the slacking phenomenon that affects robot therapy. Design/methodology/approach – The evaluation of the AC index is based on a novel technique of assistance which does not use constant or elastic forces but trains of small force impulses, with amplitude adapted to the level of impairment and a frequency of 2 Hz, which is suggested by recent results in the field of intermittent motor control. A preliminary feasibility test of the proposed method was carried out during a haptic reaching task in the absence of visual feedback, for a group of five stroke patients and an equal group of healthy subjects. Findings – The AC index appears to be stable and sensitive to training in both populations of subjects. Originality/value – The main original element of this study is the proposal of the new AC index of voluntary control associated with the new method of pulsed haptic interaction.