Skip to main content

Elbow exoskeleton mechanism for multistage poststroke rehabilitation.

Manna, S., 2020. Elbow exoskeleton mechanism for multistage poststroke rehabilitation. Doctoral Thesis (Doctoral). Bournemouth University.

Full text available as:

MANNA, Soumya Kanti_Ph.D._2020.pdf



More than three million people are suffering from stroke in England. The process of post-stroke rehabilitation consists of a series of biomechanical exercises- controlled joint movement in acute phase; external assistance in the mid phase; and variable levels of resistance in the last phase. Post-stroke rehabilitation performed by physiotherapist has many limitations including cost, time, repeatability and intensity of exercises. Although a large variety of arm exoskeletons have been developed in the last two decades to substitute the conventional exercises provided by physiotherapist, most of these systems have limitations with structural configuration, sensory data acquisition and control architecture. It is still difficult to facilitate multistage post-stroke rehabilitation to patients sited around hospital bed without expert intervention. To support this, a framework for elbow exoskeleton has been developed that is portable and has the potential to offer all three types of exercises (external force, assistive and resistive) in a single structure. The design enhances torque to weight ratio compared to joint based actuation systems. The structural lengths of the exoskeleton are determined based on the mean anthropometric parameters of healthy users and the lengths of upperarm and forearm are determined to fit a wide range of users. The operation of the exoskeleton is divided into three regions where each type of exercise can be served in a specific way depending on the requirements of users. Electric motor provides power in the first region of operation whereas spring based assistive force is used in the second region and spring based resistive force is applied in the third region. This design concept provides an engineering solution of integrating three phases of post-stroke exercises in a single device. With this strategy, the energy source is only used in the first region to power the motor whereas the other two modes of exercise can work on the stored energy of springs. All these operations are controlled by a single motor and the maximum torque of the motor required is only 5 Nm. However, due to mechanical advantage, the exoskeleton can provide the joint torque up to 10 Nm. To remove the dependency on biosensor, the exoskeleton has been designed with a hardware-based mechanism that can provide assistive and resistive force. All exoskeleton components are integrated into a microcontroller-based circuit for measuring three joint parameters (angle, velocity and torque) and for controlling exercises. A user-friendly, multi-purpose graphical interface has been developed for participants to control the mode of exercise and it can be managed manually or in automatic mode. To validate the conceptual design, a prototype of the exoskeleton has been developed and it has been tested with healthy subjects. The generated assistive torque can be varied up to 0.037 Nm whereas resistive torque can be varied up to 0.057 Nm. The mass of the exoskeleton is approximately 1.8 kg. Two comparative studies have been performed to assess the measurement accuracy of the exoskeleton. In the first study, data collected from two healthy participants after using the exoskeleton and Kinect sensor by keeping Kinect sensor as reference. The mean measurement errors in joint angle are within 5.18 % for participant 1 and 1.66% for participant 2; the errors in torque measurement are within 8.48% and 7.93% respectively. In the next study, the repeatability of joint measurement by exoskeleton is analysed. The exoskeleton has been used by three healthy users in two rotation cycles. It shows a strong correction (correlation coefficient: 0.99) between two consecutive joint angle measurements and standard deviation is calculated to determine the error margin which comes under acceptable range (maximum: 8.897). The research embodied in this thesis presents a design framework of a portable exoskeleton model for providing three modes of exercises, which could provide a potential solution for all stages of post- stroke rehabilitation.

Item Type:Thesis (Doctoral)
Additional Information:If you feel that this work infringes your copyright please contact the BURO Manager.
Uncontrolled Keywords:bio-medical; medical-robotics; exoskeleton; mechanism; stroke rehabilitation
Group:Faculty of Science & Technology
ID Code:34157
Deposited By: Unnamed user with email symplectic@symplectic
Deposited On:16 Jun 2020 15:23
Last Modified:16 Jun 2020 15:23


Downloads per month over past year

More statistics for this item...
Repository Staff Only -