Recovering Planned Trajectories in Robotic Rehabilitation Therapies under the Effect of Disturbances

Recovering Planned Trajectories in Robotic Rehabilitation Therapies under the Effect of Disturbances

Vijaykumar Rajasekaran (Institute for Bioengineering of Catalonia & Universitat Politècnica de Catalunya, Barcelona Tech, Barcelona, Spain), Joan Aranda (Institute for Bioengineering of Catalonia & Universitat Politècnica de Catalunya, Barcelona Tech, Barcelona, Spain) and Alicia Casals (Institute for Bioengineering of Catalonia & Universitat Politècnica de Catalunya, Barcelona Tech, Barcelona, Spain)
Copyright: © 2014 |Pages: 16
DOI: 10.4018/ijsda.2014040103
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Robotic rehabilitation is an emerging technology in the field of Neurorehabilitation, which aims to achieve an effective patient recovery. This research focusses on the control strategy for an assistive exoskeleton aiming to reduce the effects of disturbances on planned trajectories during rehabilitation therapies. Disturbances are mostly caused by muscle synergies or by unpredictable actions produced by functional electrical stimulation. The effect of these disturbances can be either assistive or resistive forces depending on the patient's movement, which increase or decrease the speed of the affected joints by forcing the control unit to act consequently. In some therapies, like gait assistance, it is also essential to maintain synchronization between joint movements, to ensure a dynamic stability. A force control approach is used for all the joints individually, while two control methods are defined to act when disturbances are detected: Cartesian position control (Cartesian level) and Variable execution speed (joint level). The trajectory to be followed by the patient is previously recorded using an active exoskeleton, H1, worn by healthy subjects. A realistic simulation model of the exoskeleton is used for testing the effect of disturbances on the particular joints and on the planned trajectory and for evaluating the performance of the two proposed control methods. The performances of the presented methods are evaluated by comparing the resulting trajectories with respect to those planned. The evaluation of the most suitable method is performed considering the following factors: stability, minimum time delay and synchronization of the joints.
Article Preview

1. Introduction

Robotics based rehabilitation therapy has become a topic of interest for many rehabilitation therapists and also for researchers working in the field of medical robotics (Azar & Eljamel, 2012). Robotic rehabilitation is not only addressed to patients training, but also to provide gait assistance to individuals with severe neurologic disorders, such as Spinal Cord Injury (SCI) or Stroke. This type of therapeutic assistance can also be extended to patients with muscle disorders or with other post-operative rehabilitation requirements. Robots are suitable for testing and application of motor learning principles to neurorehabilitation. Studies on the effectiveness of robotic neurorehabilitation have proven that robots are beneficial in measuring the patient’s impairment level, but they have not demonstrated to be so effective regarding functional outcomes (Huang & Krakauer, 2009). Robots have also been used extensively to study the process of motor learning in healthy subjects. For this study (Shadmehr & Wise, 2007), a force perturbation, which causes large errors in the trajectory, is applied to the robot and the subjects must adapt to it. This process of inducing perturbations has enabled scientists to test several hypotheses to study the computational mechanisms of motor control and motor learning. An analytical study by Dollar and Herr (2008), explains the classification of exoskeletons based on how they are applied: performance based or active exoskeletons. Performance enhancing exoskeletons help in improving locomotion, reducing musculoskeletal forces and muscle fatigue and improving bipedal stability. Active orthosis are capable of augmenting and controlling the power at the joints according to the nature of the disability. Many active exoskeletons have been developed for active gait restoration with considerable variations in actuation and sensing technologies and in control strategies. However, there are still some limitations to overcome in providing effective gait compensation (Waldner et al., 2008).

Nef et al (2007) developed a patient- robot cooperative therapy, which aims to improve rehabilitation and therapeutic progress of stroke and SCI patients by encouraging more intense training and increasing the patient’s motivation and activity using an upper limb rehabilitation robot ARMin. For lower limbs, a therapeutic robot Physiotherabot has been developed to perform exercises with the help of a human-machine interface, which operates even in the absence of physiotherapists. This robot is capable of transforming its joint configuration based on the feedback data received from the patient (Erhan & Adli, 2011). As a performance enhancer, hybrid control of the augmentative robot BLEEX (Kazerooni et al., 2006) using a load carrying lower extremity exoskeleton has demonstrated to offer robustness to the changing backpack payload (18kg/40lbs) dynamics. A sensitivity amplification controller is used for the swing leg and hybrid control for the stance leg. All these studies imply the need of an efficient and effective therapy by considering the internal and external changes affecting the therapy or movement. The efficiency and effectiveness of a system can be ensured by the reactive and proactive behavior of the robot with respect to the patient’s movement. The present work demonstrates a patient-robot cooperative therapy for a wearable lower limb exoskeleton in the presence of disturbances. A reactive or proactive behavior of a system can be demonstrated by implementing an efficient control strategy.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing