Title: Force-position control of a robotic exoskeleton to provide upper extremity movement assistance

Authors: Mohammad Habibur Rahman; Cristóbal Ochoa-Luna; Md Jahidur Rahman; Maarouf Saad; Philippe Archambault

Addresses: Department of Electrical Engineering, École de Technologie Supérieure, 1100 Notre-Dame, Montréal, H3C 1K3, Canada; School of Physical and Occupational Therapy, McGill University, 845 Sherbrooke St. W, Montreal, QC H3A 0G4, Canada ' Department of Electrical Engineering, École de Technologie Supérieure, 1100 Notre-Dame, Montréal, H3C 1K3, Canada ' Department of Electrical Engineering, École de Technologie Supérieure, 1100 Notre-Dame, Montréal, H3C 1K3, Canada ' Department of Electrical Engineering, École de Technologie Supérieure, 1100 Notre-Dame, Montréal, H3C 1K3, Canada ' School of Physical and Occupational Therapy, McGill University, 845 Sherbrooke St. W, Montreal, QC H3A 0G4, Canada; Interdisciplinary Research Center in Rehabilitation (CRIR), 2275 Avenue Laurier E, Montréal, QC, H2H 2N8, Canada

Abstract: This paper presents an upper extremity (UE) wearable robot, ETS-MARSE and its control strategy to provide movement assistance and active rehabilitation exercises to physically disabled individuals having impaired UE function. The ETS-MARSE was designed to be worn on the lateral side of UE and is able to assist arm movements at the level of shoulder, elbow, forearm and wrist joint movements. Considering the dynamic modelling of the robot system and the UE motion which are non-linear in nature, a non-linear sliding mode control with exponential reaching law was used to manoeuvre the ETS-MARSE and to provide both passive and assisted arm movement therapy. To provide passive arm movement therapy, pre-programmed trajectories corresponding to recommended passive rehabilitation exercises were used to manoeuvre the ETS-MARSE, whereas in case of assisted rehabilitation therapy user interaction wrist force sensor signals were used to steer the ETS-MARSE in assisting the UE movements. Experiments involving healthy human subjects were performed with the developed ETS-MARSE to evaluate the controller's performance and that of the ETS-MARSE in regards to providing the therapeutic exercises. Experimental results indicate that with the proposed control strategy, ETS-MARSE can effectively deliver rehabilitation exercises.

Keywords: force-position transformer; exoskeleton robots; nonlinear SMC; sliding mode control; motion assistance; assisted rehabilitation therapy; upper extremity impairment; ETS-MARSE; wearable robots; robot control; arm movements; dynamic modelling; wrist force sensors; rehabilitation exercises; robot trajectories; robotic exoskeletons.

DOI: 10.1504/IJMIC.2014.062026

International Journal of Modelling, Identification and Control, 2014 Vol.21 No.4, pp.390 - 400

Published online: 23 Aug 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article