Beom-Chan Lee

A Wearable Device for Real-Time Motion Error Detection and Vibrotactile Instructional Cuing

We have developed a mobile instrument for motion instruction and correction (MIMIC) that enables an expert (i.e., physical therapist) to map his/her movements to a trainee (i.e., patient) in a hands-free fashion. MIMIC comprises an expert module (EM) and a trainee module (TM). Both the EM and TM are composed of six-degree-of-freedom inertial measurement units, microcontrollers, and batteries. The TM also has an array of actuators that provide the user with vibrotactile instructional cues. The expert wears the EM, and his/her relevant body position is computed by an algorithm based on an extended Kalman filter that provides asymptotic state estimation. The captured expert body motion information is transmitted wirelessly to the trainee, and based on the computed difference between the expert and trainee motion, directional instructions are displayed via vibrotactile stimulation to the skin. The trainee is instructed to move in the direction of the vibration sensation until the vibration is eliminated. Two proof-of-concept studies involving young, healthy subjects were conducted using a simplified version of the MIMIC system (pre-specified target trajectories representing ideal expert movements and only two actuators) during anterior-posterior trunk movements. The first study was designed to investigate the effects of changing the expert-trainee error thresholds (0.5°, 1.0°, and 1.5°) and varying the nature of the control signal (proportional, proportional plus derivative). Expert-subject cross-correlation values were maximized (0.99) and average position errors (0.33°) and time delays (0.2 s) were minimized when the controller used a 0.5° error threshold and proportional plus derivative feedback control signal. The second study used the best performing activation threshold and control signal determined from the first study to investigate subject performance when the motion task complexity and speed were varied. Subject performance decreased as motion speed and complexity increased.

Immersive live sports experience with vibrotactile sensation

This paper presents a vibrotactile display system designed with an aim of providing immersive live sports experience. Preliminary user studies showed that with this display subjects were 35% more accurate in interpreting an ambiguous visual stimulus showing a ball either entering or narrowly missing a football net. About 80% of subjects could judge the correct ball paths in the presences of ambiguous visual stimuli. Without the tactile display, only 60% correct paths are judged from the visual display.

Postural Reorganization Induced by Torso Cutaneous Covibration

Cutaneous information from joints has been attributed proprioceptive properties similar to those of muscle spindles. This study aimed to assess whether vibration-induced changes in torso cutaneous information contribute to whole-body postural reorganization in humans. Ten healthy young adults stood in normal and Romberg stances with six vibrating actuators positioned on the torso in contact with the skin over the left and right external oblique, internal oblique, and erector spinae muscle locations at the L4/L5 vertebrae level. Vibrations around the torso were randomly applied at two locations simultaneously (covibration) or at all locations simultaneously. Kinematic analysis of the body segments indicated that covibration applied to the skin over the internal oblique muscles induced shifts of both the head and torso in the anterior direction (torso flexion) while the hips shifted in the posterior direction (ankle plantar flexion). Conversely, covibration applied to the skin over the erector spinae muscle locations produced opposite effects. However, covibration applied to the skin over the left internal oblique and left erector spinae, the right internal oblique and right erector spinae, or at all locations simultaneously did not induce any significant postural changes. In addition, the center of pressure position as measured by the force plate was unaffected by all covibration conditions tested. These results were independent of stance and suggest an integrated and coordinated reorganization of posture in response to vibration-induced changes in cutaneous information. In addition, combinations of vibrotactile stimuli over multiple locations exhibit directional summation properties in contrast to the individual responses we observed in our previous work.

Effects of aging and tactile stochastic resonance on postural performance and postural control in a sensory conflict task

Abstract Postural control in certain situations depends on functioning of tactile or proprioceptive receptors and their respective dynamic integration. Loss of sensory functioning can lead to increased risk of falls in challenging postural tasks, especially in older adults. Stochastic resonance, a concept describing better function of systems with addition of optimal levels of noise, has shown to be beneficial for balance performance in certain populations and simple postural tasks. In this study, we tested the effects of aging and a tactile stochastic resonance stimulus (TSRS) on balance of adults in a sensory conflict task. Nineteen older (71–84 years of age) and younger participants (22–29 years of age) stood on a force plate for repeated trials of 20 s duration, while foot sole stimulation was either turned on or off, and the visual surrounding was sway-referenced. Balance performance was evaluated by computing an Equilibrium Score (ES) and anterior–posterior sway path length (APPlength). For postural control evaluation, strategy scores and approximate entropy (ApEn) were computed. Repeated-measures ANOVA, Wilcoxon signed-rank tests, and Mann–Whitney U-tests were conducted for statistical analysis. Our results showed that balance performance differed between older and younger adults as indicated by ES (p = 0.01) and APPlength (0.01), and addition of vibration only improved performance in the older group significantly (p = 0.012). Strategy scores differed between both age groups, whereas vibration only affected the older group (p = 0.025). Our results indicate that aging affects specific postural outcomes and that TSRS is beneficial for older adults in a visual sensory conflict task, but more research is needed to investigate the effectiveness in individuals with more severe balance problems, for example, due to neuropathy.

Comparison of non-volitional postural responses induced by two types of torso based vibrotactile stimulations

The purpose of this study was to characterize the non-volitional postural responses to torso-based vibrotactile stimulation as a function of stimulation location for two types of vibrating actuators (tactors). Eleven young healthy adults were asked to maintain an upright erect posture with their eyes closed. Two types of tactors, Tactaid (electromagnetic inertial transducer) and C-2 (voice-coil-type linear transducer), were placed over the left and right external oblique, internal oblique, and erector spinae muscles in two different trial series. Regardless of the tactor type, vibration applied over the internal oblique and erector spinae muscles induced a postural shift in the direction of the stimulation. For these four locations, the root-mean-square (RMS) of the sway was significantly greater during vibration than immediately before or after stimulation. Vibration-induced postural shifts and increases in RMS sway were greater for the C-2 than Tactaid tactors. Simultaneous activation of all tactors or those over the external oblique muscles did not produce significant directional postural shifts or increases in sway, regardless of the tactor type. The directional shifts of posture suggest that these non-volitional responses should be considered to improve the use of torso-based vibrotactile sensory augmentation display designed for clinical balance applications.

Teleoperation of a multi-purpose robot over the internet using augmented reality

Bilateral teleoperation using augmented reality is proposed for a multi-purpose robot called SpiderBot-II that is an indoor-installed wire-driven parallel manipulator. It is intended to be used for various applications, including in-house rehabilitation training and daily life assistance such as walking assistance and health monitoring, especially for the elderly or the handicapped that spends most of time at home. Through the teleoperation over the Internet, a therapist or an attendant in a remote site can give help to the physically disabled by directly manipulating the robot. For easy recognition of the obstacle, predefined markers are attached to each obstacle in the workspace. For better teleoperation, moreover, reaction force between obstacles and the end-effector, which is calculated using force field, is given to a remote operator and this enables the operator to perform the teleoperation more effectively. Force field, which is proportional to proximity between obstacles and the end-effector, is generated to facilitate the obstacle avoidance and visually augmented on the operators screen for better recognition of obstacles.

Haptic based gait rehabilitation system for stroke patients

Among most existing gait rehabilitation robots, it is difficult to find adequate devices for gait rehabilitation of chronic stroke patients who can already stand and move but still need to rehabilitate the affected lower limb through simple, compact, and easy-to use devices. This paper presents a novel haptic based gait rehabilitation system (HGRS) which has the potential to provide over-ground gait training regimens for post-stroke ambulatory subjects. It consists of a portable cane for kinesthetic sensing and a wearable vibrotactor array for tactile biofeedback. Contact of user with the handle provides light grip force, it serves the purpose of balance assurance and increased muscle activity through light touch concept and vibrotactors contribute in enhancing the gait modification through afferent signal of vibration. Walking trials conducted with stroke patients indicate increased muscle activation and balance, and improved temporal symmetry with use of HGRS. HGRS is capable of assisting physical therapists in training individuals with stroke suffering from gait abnormalities. In addition, it is easy to use and low-cost which makes it reachable to a vast domain of subjects suffering from gait abnormalities.

Smartphone based fall detection system

This paper describes the design of a smartphone based fall detection system and characterizes the preliminary efficacy of the proposed system in activities of daily living (ADLs). Using the embedded sensors available in a smartphone (i.e., accelerometer, gyroscope and magnetometer), kinematic analysis of movement can be performed in real-time, allowing for continuous monitoring of fall status. Fall sensing thresholds are defined based on angular rate of change (TH1), maximum acceleration (TH2), and maximum attitude change (TH3). TH1 is measured from the resultant pitch and roll angular velocity vector and defined as 3.1 rad/s (~180°/s). TH2 is measured from the resultant acceleration vector and defined as 1.6 g. TH3 is measured from the resultant vector of the pitch and roll angles, and defined at 0.59 rad (39°). A proof-of-concept study was performed on five ADL tasks: 1) comfortable walking, 2) stand-to-seated posture, 3) seated-to-standing posture, 4) pivoting at the waist to pick up an object, and 5) stand-to-seated-to-laying transition. No trials violated the defined thresholds for fall detection, signifying no false positives. These results are important for the definition of machine learning algorithms, currently under development, to minimize false positive and false negative fall detection events.

Effects of attractive versus repulsive vibrotactile instructional cues during motion replication tasks

The Mobile Instrument for Motion Instruction and Correction (MIMIC) enables an expert (i.e., physical therapist) to map his/her movements to a trainee (i.e., patient) in a hands-free fashion. MIMIC comprises an Expert Module (EM) and a Trainee Module (TM); both modules include six-degree-of-freedom inertial measurement units, microcontrollers, and batteries. The TM also includes actuators that provide the trainee with vibrotactile instructional cues. The estimated expert body motion information is transmitted wirelessly to the trainee; based on the computed difference between the motions of the expert and trainee, directional instructions are displayed to the trainees skin via vibrotactile stimulation. This study examined anterior-posterior trunk movements using a simplified version of the MIMIC system in which only two actuators were used to provide feedback and pre-recorded target trajectories were used to represent ideal expert movements. The study was designed to investigate the effects of attractive versus repulsive vibrotactile instructional cues when the motion speed and task complexity were varied. Preliminary results (n = 12) suggest that repulsive vibrotactile instructional cues lead to the greatest correlation between expert and subject motion, the least time delay, and the least tilt error.

Smooth haptic interaction in broadcasted augmented reality

This paper presents smooth haptic interaction methods for an immersive and interactive broadcasting system combining haptics in augmented reality. When touching the broadcasted augmented virtual objects in the captured real scene, problems of force trembling and discontinuity occur due to static registration errors and slow marker pose update rate, respectively. In order to solve these problems, threshold and interpolation methods are proposed respectively. The resultant haptic interaction provides smoother continuous tremble-free force sensation.