Ozer Unluhisarcikli

Design and control of a robotic lower extremity exoskeleton for gait rehabilitation

Design and control of an active knee rehabilitation orthotic system called ANdROS that was designed as a wearable and portable gait rehabilitation tool is presented. A corrective force field that reinforces a desired gait pattern is applied to the patients impaired leg around the knee joint via an impedance controlled exoskeleton. The impedance controller is synchronized with the patients walking phase which is estimated from the kinematic measurements of the healthy leg. The performance of the controller is evaluated through bench-testing.

A robotic hand rehabilitation system with interactive gaming using novel Electro-Rheological Fluid based actuators

A newly developed hand rehabilitation system is presented that combines robotics and interactive gaming to facilitate repetitive performance of task specific exercises for patients recovering from neurological motor deficits. A two degree of freedom robotic interface allows coordinated motions of the forearm and the hand (pronation/supination and grasp/release, respectively). It is driven by two novel Electro-Rheological Fluid based hydraulic actuators. Tests were conducted to characterize these actuators, and feed-forward controllers were developed for their force/torque control. A virtual reality environment (maze game) was developed in which the robot applies force fields to the user as the user navigates the environment, forming a haptic interface between the patient and the game.

Variable Resistance Hand Device using an electro-rheological fluid damper

This paper presents the design, fabrication, control and testing of the third generation prototype of a novel, one degree of freedom (DOF) Variable Resistance Hand Device (VRHD) that was designed for isotonic, isokinetic, and variable resistance grasp and release exercises. Its principle functionality is derived from an electro-rheological fluid based controllable damper that allows continuously variable modulation of dynamic resistance throughout its stroke. The VRHD system consists of the patient actuated device, the control electronics and software, the practitioner graphical interface and the patients virtual reality game software. VRHD was designed and experimentally shown to be fully Magnetic Resonance Imaging (MRI) compatible so that it can be used in brain MR imaging during handgrip rehabilitation.

Haptic system for hand rehabilitation integrating an interactive game with an advanced robotic device

A haptic system for hand rehabilitation is presented that combines robotics and interactive virtual reality to facilitate repetitive performance of task specific exercises for patients recovering from neurological motor deficits. A two degree of freedom robotic interface allows coordinated motions of the forearm and the hand (pronation/supination and grasp/release, respectively). It is driven by two novel Electro-Rheological Fluid (ERF) based hydraulic actuators. Tests were conducted to characterize these actuators, and feed-forward controllers were developed for their force/torque control. A virtual reality environment (maze game) was developed in which the robot applies force fields to the user as the user navigates the environment, forming a haptic interface between the patient and the game. Proof of concept testing was performed on the virtual environment to analyze the use interaction of haptic feedback and a virtual game.

Design of human — Machine interface and altering of pelvic obliquity with RGR Trainer

The Robotic Gait Rehabilitation (RGR) Trainer targets secondary gait deviations in stroke survivors undergoing rehabilitation. Using an impedance control strategy and a linear electromagnetic actuator, the device generates a force field to control pelvic obliquity through a Human-Machine Interface (i.e. a lower body exoskeleton). Herein we describe the design of the RGR Trainer Human-Machine Interface (HMI) and we demonstrate the systems ability to alter the pattern of movement of the pelvis during gait in a healthy subject. Results are shown for experiments during which we induced hip-hiking — in healthy subjects. Our findings indicate that the RGR Trainer has the ability of affecting pelvic obliquity during gait. Furthermore, we provide preliminary evidence of short-term retention of the modified pelvic obliquity pattern induced by the RGR Trainer.

Robotic Systems for Gait Rehabilitation

Human walking is impaired by various neurological diseases such as stroke. Gait restoration is a major goal in neurological rehabilitation following stroke. Robotic devices have been developed to assist locomotor training improving gait function and thus a stroke survivor’s independence. This chapter presents robotic systems that have been developed specifically for gait rehabilitation providing pelvic, hip and/or knee motion assistance. Robotic systems allow clinicians to increase the duration, intensity and specificity of treatment compared to traditional physical therapy. These factors could result in a faster and increased level of recovery of functional capability thus leading to an improvement of patient’s level of independence and quality of life. In addition, robotic systems could be used to reduce the number of physical therapists involved in the treatment of each patient. In fact, there is great interest in robot-assisted rehabilitation for partially automating such therapy, to enable just one physical therapist to administer gait training instead of at least two. A major limitation in the use of robotic systems for gait rehabilitation is their high cost. Currently, commercially available robotic solutions for automation of gait rehabilitation physical therapy cost between

PHASE-DEPENDENT KNEE MODEL IDENTIFICATION IN GAIT MOTION USING AN ACTIVE KNEE REHABILITATION ORTHOTIC DEVICE

60,000 and

Lower Extremity Exoskeleton for Gait Retraining

300,000 and thus very few clinical facilities can afford them. In addition, low cost robotic devices for gait rehabilitation could be used by post-stroke survivors at home, in order to accelerate and intensify the rehabilitation process and improve the therapeutic outcomes.

ROBOTIC GAIT REHABILITATION TRAINING SYSTEM WITH ORTHOPEDIC LOWER BODY EXOSKELETON FOR TORQUE TRANSFER TO CONTROL ROTATION OF PELVIS DURING GAIT

This paper presents a new approach in modeling human motions using robotic rehabilitation devices. A phase dependent identification method is proposed that is based on applying a Gaussian white noise around the knee joint through the use of a rehabilitation robotic exoskeleton. The input-output relationship is analyzed and the concept to represent the walking motion as a phase-dependent action is described. The method takes advantage of the time independent property of Gaussian noise and shows great accuracy in modeling human limb motions when compared to classical (i.e. global) models. The new modeling method for human limb motions is important in the control and planning of rehabilitation robotic devices and in evaluating human biomechanics using a portable gait measuring device.