Development of a whole arm wearable robotic exoskeleton for rehabilitation and to assist upper limb movements

Totally useless until this is tested out in stroke survivors

 Development of a whole arm wearable robotic exoskeleton for rehabilitation and to assist upper limb movements

M. H. Rahman
†
,
‡
∗
, M. J. Rahman
†
, O. L. Cristobal
†
,M. Saad
†
, J. P. Kenn´e
†
 and P. S. Archambault
‡
,
§
†
 Department of Electrical Engineering, ´  Ecole de Technologie Sup´ erieure (ETS), Montr ´ eal, Canada
‡
School of Physical & Occupational Therapy, McGill University, Montr ´ eal, Canada
§
 Interdisciplinary Research Center in Rehabilitation (CRIR), Montr ´ eal, Canada
(Accepted December 22, 2013)SUMMARY
To assist physically disabled people with impaired upper limb function, we have developed a new7-DOF exoskeleton-type robot named
 Motion Assistive Robotic-Exoskeleton for Superior Extremity(
 ETS-MARSE f46) to ease daily upper limb movements and to provide effective rehabilitation therapy to the superior extremity. The ETS-MARSE comprises a shoulder motion support part, an elbow and forearm motion support part, and a wrist motion support part. It is designed to be worn on the lateral side of the upper limb in order to provide naturalistic movements of the shoulder (vertical and horizontal flexion/extension and internal/external rotation), elbow (flexion/extension), forearm(pronation/supination), and wrist joint (radial/ulnar deviation and flexion/extension). This paper focuses on the modeling, design, development, and control of the ETS-MARSE. Experiments were carried out with healthy male human subjects in whom trajectory tracking in the form of passive rehabilitation exercises (i.e., pre-programmed trajectories recommended by a therapist/clinician)were carried out. Experimental results show that the ETS-MARSE can efficiently perform passive rehabilitation therapy.KEYWORDS: Robotic exoskeleton; Nonlinear control; Physical disability; Passive rehabilitation;Upper limb impairment.¹ 

Introduction

Upper limb impairment(such as full or partial loss of function in shoulder joint, elbow joint, and wrist joint movements) is very common in the elderly, but can also be a secondary effect due to strokes,cardiovascular diseases, trauma, sports injuries, occupational injuries, and spinal cord injuries. A proper functioning of the upper limb is very important for the performance of essential daily activities. According to the World Health Organization, each year strokes and cardiovascular diseases affect more than 15 million people worldwide.⁵Of these, 85% of stroke survivors incur acute arm impairment,and 40% are chronically impaired or permanently disabled, thereby placing burden on the family and community.⁸Rehabilitation programs are the main method to promote functional recovery in these subjects.¹¹Since the number of such cases is constantly growing, and the treatment duration is long,requiring skilled therapists or clinicians, introducing robots could therefore significantly contribute to the success of these programs in providing very efficient passive and tireless rehabilitation for long a period of time as the proposed
 Motion Assistive Robotic-Exoskeleton for Superior Extremity ( ETS-MARSE) demonstrates.It has been shown in several studies that intensive and repetitive therapies significantly improve motor skills.¹³Further studies have revealed that enhanced motor learning occurs when patients  Development of a whole arm wearable roboticexoskeleton for rehabilitation and to assist upper limb movements
 practice a variety of functional tasks^14–18(such as reaching movements) and receive feedback intermittently (e.g., visual and haptic feedback in virtual reality).^{9,15–18,20–22}Therefore, these key factors of therapy are to be integrated in rehabilitation paradigms, and this can be done through rehabilitation robotics. Moreover, recent studies also reveal that robot-aided therapy and virtual reality based rehabilitation significantly reduce arm impairment^{9,15–18,20–23}and improve motor function, allowing the subject to regain upper limb function of motion.^24,25To assist physically weakened and/or disabled individuals with impaired upper limb function,extensive research has been carried out in many branches of robotics, particularly on wearable robot(e.g., exoskeletons, powered orthotic devices, etc.) and/or end effector based robotic devices (i.e.,devices which do not actively support or hold the subject’s arm but connect with the subject’s hand or forearm).^24,26–28The exoskeleton-type robotic devices found in literature are either chair-^30,31or floor-mounted,^19,29,32but the end-effector devices are commonly found as floor-/desk-mounted. TableI highlights and compares some features(e.g., degrees of freedom(DOFs), sensors and actuators used, placement of actuators, actuation mechanism, therapeutic regime) of these devices.Although much progress has been made in robotics, we are still far from the desired objective, as existing robots are not yet able to restore bodily mobility or function. This is due to limitations in the area of proper hardware design and that of control algorithms in terms of developing intelligent and autonomous robots that perform intelligent tasks. Some of the notable hardware limitations in the existing exoskeleton systems include limited degrees of freedom and range of motion^3,9,27,31as com-pared with that of human upper extremities, robust and complex structures,¹²weak joint mechanismsof the exoskeleton system,^30,31bulky actuated joints,¹⁷relatively heavy weight of the exoskeletonarm,^17,32lack of proper safety measures and compensation for gravity forces,^12,26,27,31and complex cable routing for transmission mechanisms.^6,9,30The ETS-MARSE developed in this research has taken the above limitations into account, and is designed based on upper limb joints movement; it has a relatively low weight, can be easily fitted or removed, and is able to effectively compensate for gravity. Moreover, to avoid the complex cable routing encountered in many exoskeleton systems,^6,9,30an innovative gear transmission mechanism has been developed for forearm pronation/supination andshoulder joint internal/external rotation. Note that cable transmission always adds some undesirablevibration and can loosen up during operation; therefore it should be avoided. Some devices used gearmechanism with a closed circular structure of forearm/upper arm cup.^19,30However, it is unrealistic and inconvenient to insert and remove the arm through a closed circular structure. Although extensive research has been carried out in developing smart rehabilitative and motion-assistive devices, a few numbers of such devices can be found in the literature that focused on thepassive rehabilitation approach. The passive arm movement therapy is considered as a first stage of physiotherapy treatment (exercise) that is usually given to the patients to improve passive range of movements (ROMs). Therefore, this therapeutic approach should be given utmost importance.Table II briefly summarizes and compares some features of existing robotic devices that focused on passive rehabilitation therapy. It can be seen from Table II that the weight of the ETS-MARSE arm from shoulder joint to wrist handle is 7.072 kg. Compared with the existing exoskeleton devices having at least shoulder and elbow motion support parts and focused on passive and/or active rehabilitation approach, the developed ETS-MARSE is found to be light in weight. For example, weight of CADEN-7 (7 DoFs, focused on passive rehabilitation, but not verified experimentally) is 9.2 kg excluding the weight of actuators. The exoskeleton was primarily developed for human power assist and use scable mechanism to transmit power to the joints. The inherited problems in cable-driven systems is already discussed. Other passive rehabilitation robotic devices, such as iHandRehab,⁴IntelliArm,⁷and Hand Motion Assist Robot,¹⁰only focused on the hand and wrist rehabilitation. Soft actuated exoskeleton,¹²though have 7 DoFs, it uses pneumatic muscle actuators, and therefore requires cumbersome experimental setup to operate. Moreover, the exoskeleton structure comprises week  joint mechanism and suffers from safety issues. Also, it can be seen from Table II that the majority of exoskeleton systems uses proportional integral and derivative(PID)-based control approach, meaning that dynamic models of the system as well as that of human upper limb were ignored. Our developed ETS-MASRE designed to provide every variety of movements to the upper extremity also addressed the control issue and has used model-based control approach.Most of the existing rehabilitative devices, although having limited degrees of freedom,demonstrated robot-assisted active rehabilitation exercises.^36,39Note that passive arm movement therapy is the very first type of physiotherapy treatment given to the subjects/patients who are unable to actively move their arms throughout their complete range of motion following a surgery
40
of the shoulder-, elbow-, or wrist joint due to joint dislocation, or as a result of a stroke mostly dueto spasticity and increased muscle tone.^42,43herefore, this issue should be properly addressed inrehabilitation robotics. As a solution to this issue, in this paper we have also presented a controlstrategy with nonlinear computed torque control technique to provide passive rehabilitation therapyfor single- and multi-joint movements.In our previous research, we developed exoskeleton-type robots for rehabilitation to assist elbow-,forearm-,
1
and wrist joint movements.²In a continuing effort toward providing movement assistanceto the whole arm (shoulder, elbow, forearm, and wrist joint) in this research, we have focused on thedevelopment of an innovative new 7-DoF exoskeleton-type robot (ETS-MARSE).The paper presents a complete bio-mechatronic system that includes (i) mechanical design of the ETS-MARSE; (ii) electrical and electronic design of the ETS-MARSE; (iii) control strategy to maneuver the ETS-MARSE; and (iv) experiment results demonstrating passive upper limb rehabilitation exercises with the developed ETS-MARSE. The ETS-MARSE was designed based on upper limb articulations and movement. Modified Denavit–Hartenberg (DH) conventions⁴⁸were used in developing the kinematic model. In dynamic modeling and control, subject parameters as well as the ETS-MARSE parameters, such as arm length,mass of different segments, robot arm link lengths, and inertia, were estimated according to the upper limb properties of a typical adult.⁴⁶The ETS-MARSE is supposed to be worn on the lateral side of the upper arm, with the aim of providing effective rehabilitation for the shoulder joint (3 DOFs:horizontal and vertical motion, flexion/extension motion, and internal/external rotation), elbow joint(1 DOF: flexion/extension motion), forearm (1 DOF: pronation/supination motion), and wrist joint(2 DOFs: flexion/extension motion and radial/ulnar deviation) movements. The entire ETS-MARSEis manufactured using aluminum, which gives the structure a relatively lightweight. Brushless DC motors incorporated with harmonic drives (HD) are used to actuate the ETS-MARSE. Note that HDs are compact in shape, low in weight, have low/zero backlash properties, and also are back-drivable.³³,⁴⁷As the ETS-MARSE will be in direct contact with a human subject (i.e., robot user),mechanical stoppers are added to each rotational joint to limit its movement within the anatomical range of the human upper limb.⁴⁶The actuation mechanisms developed for the shoulder joint’s internal/external rotation (1 DOF)support part and the forearm motion support part (1 DOF) are somewhat complex as it is practically impossible to place any actuator along the axis of rotation of the upper arm (e.g., with the humerus/radius) due to the anatomical configuration of the human arm. Moreover, as the ETS-MARSE will be worn on the lateral side of the arm, there will be an offset between the upper arm axis of rotation and the actuator axis of rotation. Although in gear mechanisms actuators can be placed at a certain offset (eccentricity) with respect to the desired axis of rotation (for instance, axis of rotation of forearm), such a mechanism is not suitable for our purposes because in this case meshing gears are supposed to rotate around a physical axis of rotation (e.g., shaft). We are unable to fit such a mechanical shaft along the line of axis of upper arm motion (e.g.,with the humerus/radius). Therefore,we have introduced and developed an alternate gear mechanism where motion is transmitted from an anti-backlash gear (mounted on a motor shaft) to an open-type, custom-made meshing ring gear attached rigidly to the upper arm cup. This gear mechanism is discussed in Section 2.2.In the next section a detailed overview of the development of the ETS-MARSE is presented. In Section 3, experimental results are presented to evaluate the performance of the ETS-MARSE with regard to passive rehabilitation, and finally in Section 4 the paper ends with the conclusion and the future works.