Hyung-Soon Park

Design of learning input shaping technique for residual vibration suppression in an industrial robot

In this paper, a practical method is proposed to suppress residual vibrations of industrial robots without a real-time estimation of vibration frequencies. Through theoretical analysis and experiments, we designed an input shaping technique (IST) for the first three axes of a six-degrees-of-freedom industrial robot. Iterative learning IST (LIST) is applied to the first axis to suppress its time-varying nonlinear residual vibration, while conventional IST is applied to the second and third axes. Experimental results show that LIST can suppress residual vibrations to a level similar to that of a time-varying IST which requires complicated real-time estimation of a dynamic model. The LIST is an attractive method for suppression of nonlinear and time-varying residual vibrations in industrial robots which perform repetitive tasks because most industrial robots have limited computing power and memory space in their controllers.

Integration of a Rehabilitation Robotic System (KARES II) with Human-Friendly Man-Machine Interaction Units

In this paper, we report some important results of design and evaluation of a wheelchair-based robotic arm system, named as KARES II (KAIST Rehabilitation Engineering Service System II), which is newly developed for the disabled. KARES II is designed in consideration of surveyed necessary tasks for the target users (that is, people with spinal cord injury). At first, we predefined twelve important tasks according to extensive interviews and questionnaires. Next, based on these tasks, all subsystems are designed, simulated and developed. A robotic arm with active compliance and intelligent visual servoing capability is developed by using cable-driven mechanism. Various kinds of human-robot interfaces are developed to provide broad range of services according to the levels of disability. Eye-mouse, shoulder/head interface, EMG signal-based control subsystems are used for this purpose. Besides, we describe the process of integration of our rehabilitation robotic system KARES II, and discuss about user trials. A mobile platform and a wheelchair platform are two main platforms in which various subsystems are installed. For a real-world application of KARES II system, we have performed user trials with six selected potential end-users (with spinal cord injury).

Gender Differences in Passive Knee Biomechanical Properties in Tibial Rotation

The anterior cruciate ligament (ACL) is the most commonly injured knee ligament with the highest incidence of injury in female athletes who participate in pivoting sports. Noncontact ACL injuries commonly occur with both internal and external tibial rotation. ACL impingement against the lateral wall of the intercondylar notch during tibial external rotation and abduction has been proposed as an injury mechanism, but few studies have evaluated in vivo gender‐specific differences in laxity and stiffness in external and internal tibial rotations. The purpose of this study was to evaluate these differences. The knees of 10 male and 10 female healthy subjects were rotated between internal and external tibial rotation with the knee at 60° of flexion. Joint laxity, stiffness, and energy loss were compared between male and female subjects. Women had higher laxity (p = 0.01), lower stiffness (p = 0.038), and higher energy loss (p = 0.008) in external tibial rotation than did men. The results suggest that women may be at greater risk of ACL injury resulting from impingement against the lateral wall of the intercondylar notch, which has been shown to be associated with external tibial rotation and abduction.

Developing a Multi-Joint Upper Limb Exoskeleton Robot for Diagnosis, Therapy, and Outcome Evaluation in Neurorehabilitation

Arm impairments in patients post stroke involve the shoulder, elbow and wrist simultaneously. It is not very clear how patients develop spasticity and reduced range of motion (ROM) at the multiple joints and the abnormal couplings among the multiple joints and the multiple degrees-of-freedom (DOF) during passive movement. It is also not clear how they lose independent control of individual joints/DOFs and coordination among the joints/DOFs during voluntary movement. An upper limb exoskeleton robot, the IntelliArm, which can control the shoulder, elbow, and wrist, was developed, aiming to support clinicians and patients with the following integrated capabilities: 1) quantitative, objective, and comprehensive multi-joint neuromechanical pre-evaluation capabilities aiding multi-joint/DOF diagnosis for individual patients; 2) strenuous and safe passive stretching of hypertonic/deformed arm for loosening up muscles/joints based on the robot-aided diagnosis; 3) (assistive/resistive) active reaching training after passive stretching for regaining/improving motor control ability; and 4) quantitative, objective, and comprehensive neuromechanical outcome evaluation at the level of individual joints/DOFs, multiple joints, and whole arm. Feasibility of the integrated capabilities was demonstrated through experiments with stroke survivors and healthy subjects.

IntelliArm: An exoskeleton for diagnosis and treatment of patients with neurological impairments

This paper presents a novel 7 (active) +2 (passive) degrees of freedom (DOF) exoskeleton to achieve effective rehabilitation of upper limbs with neurological impairment. The 7+2 DOF robot was designed for allowing anatomically correct motions in the upper extremity, and for measuring 18 axis forces/torques and 9 DOF positions. A novel four-step integrated rehabilitation approach was developed and tested including (1) accurate and quantitative diagnosis which is not practical to do during manual clinical examinations or with existing robotic devices; (2) strenuous stretching of spastic joints based on the diagnosis and with intelligent control to adjust the stretching velocity constantly according to the joint conditions; (3) voluntary movement exercise to perform realistic functional tasks based on the patient-specific diagnosis, and (4) outcome evaluations which assessed the treatment outcome quantitatively with multiple physiological measures.

A Portable Telerehabilitation System for Remote Evaluations of Impaired Elbows in Neurological Disorders

A portable teleassessment system was designed for remote evaluation of elbow impairments in patients with neurological disorders. A master device and a slave device were used to drive a mannequin arm and the patients arm, respectively. The elbow flexion angle and torque were measured at both the master and slave devices, and sent to each other for teleoperation. To evaluate spasticity/contracture of the patients elbow remotely, the clinician asked the patient to relax the elbow, moved the mannequin arm at a selected velocity, and haptically felt the resistance from the patients elbow. In other tasks, the patient moved his/her elbow voluntarily and the clinician observed the corresponding mannequin arm movement and determined the active range of motion (ROM). The clinician could also remotely resist the patients movement and evaluate the muscle strength. To minimize the effect of network latency, two different teleoperation schemes were used depending on the speed of the tasks. For slow movement tasks, real-time teleoperations were performed using control architectures that considered causality of the tasks, with performance similar to that during an in-person examination. For tasks involving fast movements, a teach-and-replay teleoperation scheme was used which provided the examiner with transparent and stable haptic feeling. Overall, the teleassessment system allowed the clinician to remotely evaluate the impaired elbow of stroke survivors, including assessment of the passive ROM, active ROM, muscle strength, velocity-dependent spasticity, and catch angle.

Development of a Biomimetic Hand Exotendon Device (BiomHED) for Restoration of Functional Hand Movement Post-Stroke

Significant functional impairment of the hand is common among stroke survivors and restoration of hand function should be prioritized during post-stroke rehabilitation. The goal of this study was to develop a novel biomimetic device to assist patients in producing complex hand movements with a limited number of actuators. The Biomimetic Hand Exoskeleton Device (BiomHED) is actuated by exotendons that mimic the geometry of the major tendons of the hand. Ten unimpaired subjects and four chronic stroke survivors participated in experiments that tested the efficacy of the system. The exotendons reproduced distinct spatial joint coordination patterns similar to their target muscle-tendon units for both subject groups. In stroke survivors, the exotendon-produced joint angular displacements were smaller, but not significantly different, than those of unimpaired subjects (\mbi p = 0.15-0.84). Even with limited use of the BiomHED, the kinematic workspace of the index finger increased by 63%-1014% in stroke survivors. The device improved the kinematics of the tip-pinch task in stroke survivors and resulted in a significant reduction in the fingertip-thumb tip distance ( 17.9 ±15.3 mm). This device is expected to enable effective “task-oriented” training of the hand post-stroke.

Developing an Intelligent Robotic Arm for Stroke Rehabilitation

Arm impairments in patients post stroke involve the shoulder, elbow and wrist simultaneously. Patients may develop spasticity and reduced range of motion (ROM) at the multiple joints with abnormal couplings between the multiple joints and between the multiple degrees of freedom (DOF). They may lose independent control of individual joints and coordination among the joints. This project is aimed at developing a whole arm intelligent rehabilitation robot capable of controlling the shoulder, elbow, and wrist individually and simultaneously while allowing trunk motions, with the following integrated features: 1) it has unique diagnostic capabilities to determine which joints and which DOFs have significant changes in the neuromechanical properties, which joints lose independent control, what are the abnormal couplings, and whether the problem is due to changes in passive muscle properties or active control capabilities; 2) based on the diagnosis, it stretches the spastic/deformed joints forcefully under intelligent control to loosen up the specific stiff joints/DOFs; 3) with the stiff joints loosened up, the patients practice voluntary functional movements with assistance from the robot to regain/improve their motor control capability; and 4) the outcome is evaluated quantitatively at the levels of individual joints, multiple joints/DOFs, and the whole arm.

Comparison of elliptical training, stationary cycling, treadmill walking and overground walking.

The extent to which therapeutic, exercise or robotic devices can maximize gait function is a major unresolved issue in neurorehabilitation. Several factors may influence gait outcomes such as similarity of the task to overground walking, degree of coordination within and across limbs, and cycle-to-cycle variability in each device. Our objective was to compare lower extremity kinematics, coordination and variability during four locomotor tasks: overground walking, treadmill walking, elliptical training and stationary cycling in 10 non-disabled adults (6 male; mean age 22.7±2.9 yrs, range 20-29). All first performed four overground walking trials at self-selected speed with mean temporal-spatial data used to pace the other conditions. Joint positions, excursions, and the Gait Deviation Index (GDI) were compared across conditions to evaluate kinematic similarity. Time-series data were correlated within and across limbs to evaluate intralimb and interlimb coordination, respectively. Variability in cadence was quantified to assess how constrained the locomotor rhythm was compared to overground walking. Treadmill walking most closely resembled overground with GDI values nearly overlapping, reinforcing its appropriateness for gait training. Cycling showed the largest GDI difference from overground, with elliptical closer but still a significant distance from all three. Cycling showed greater hip reciprocation Cycling and elliptical showed stronger intralimb synergism at the hip and knee than the other two. Based on kinematics, results suggest that elliptical training may have greater transfer to overground walking than cycling and cycling may be more useful for enhancing reciprocal coordination. Further evaluation of these devices in neurological gait disorders is needed.

Prefrontal, posterior parietal and sensorimotor network activity underlying speed control during walking

Accumulating evidence suggests cortical circuits may contribute to control of human locomotion. Here, noninvasive electroencephalography (EEG) recorded from able-bodied volunteers during a novel treadmill walking paradigm was used to assess neural correlates of walking. A systematic processing method, including a recently developed subspace reconstruction algorithm, reduced movement-related EEG artifact prior to independent component analysis and dipole source localization. We quantified cortical activity while participants tracked slow and fast target speeds across two treadmill conditions: an active mode that adjusted belt speed based on user movements and a passive mode reflecting a typical treadmill. Our results reveal frequency specific, multi-focal task related changes in cortical oscillations elicited by active walking. Low γ band power, localized to the prefrontal and posterior parietal cortices, was significantly increased during double support and early swing phases, critical points in the gait cycle since the active controller adjusted speed based on pelvis position and swing foot velocity. These phasic γ band synchronizations provide evidence that prefrontal and posterior parietal networks, previously implicated in visuo-spatial and somotosensory integration, are engaged to enhance lower limb control during gait. Sustained μ and β band desynchronization within sensorimotor cortex, a neural correlate for movement, was observed during walking thereby validating our methods for isolating cortical activity. Our results also demonstrate the utility of EEG recorded during locomotion for probing the multi-regional cortical networks which underpin its execution. For example, the cortical network engagement elicited by the active treadmill suggests that it may enhance neuroplasticity for more effective motor training.