Browsing by Author "Kim, Sunwook"

Now showing 1 - 20 of 38
  • Thumbnail Image
    Changes in forklift driving performance and postures among novices resulting from training using a high-fidelity virtual reality simulator: An exploratory study
    Islam, Md Shafiqul; Zahabi, Saman Jamshid Nezhad; Kim, Sunwook; Lau, Nathan; Nussbaum, Maury A.; Lim, Sol (Elsevier, 2024-11-01)
    Virtual reality (VR) has emerged as a promising tool for training. Our study focused on training for forklift driving, to address an ongoing worker shortage, and the unknown impact of repeated VR training on task performance and kinematic adaptations. We trained 20 novice participants using a VR forklift simulator over two days, with two trials on each day, and including three different driving lessons of varying difficulties. Driving performance was assessed using task completion time, and we quantified kinematics of the head, shoulder, and lumbar spine. Repeated training reduced task completion time (up to ∼29.8% of initial trial) and decreased both kinematic variability and peak range of motion, though these effects were larger for lessons requiring higher precision than simple driving maneuvers. Our results highlight the potential of VR as an effective training environment for novice drivers and suggest that monitoring kinematics could help track skill acquisition during such training.
  • Thumbnail Image
    Changes in kinematics and muscle activity when learning to use a whole-body powered exoskeleton for stationary load handling
    Park, Hanjun; Kim, Sunwook; Nussbaum, Maury A.; Srinivasan, Divya (SAGE, 2022-10-11)
  • Thumbnail Image
    Changes in lower-limb joint torques when using a passive back-support exoskeleton for level walking
    Park, Jang-Ho; Kim, Sunwook; Nussbaum, Maury A.; Srinivasan, Divya (SAGE, 2021-09)
  • Cognitive Workload of Novice Forklift Truck Drivers in VR-based Training
    Jamshid Nezhad Zahabi, Saman; Shafiqul Islam, Md; Kim, Sunwook; Lau, Nathan; Nussbaum, Maury A.; Lim, Sol (SAGE, 2023-10-19)
    There is increasing use of Virtual Reality (VR) to train forklift truck operators but a lack of sufficient understanding of how cognitive workload changes with respect to different task demands in VR-based training. In this study, 19 novice participants completed three forklift driving lessons with varying difficulty levels (low, medium, and high) using a VR simulator. To examine the effect of repeated training on cognitive workload, two sessions were repeated by participants using the same procedures. Cognitive workload was assessed with objective (electroencephalogram [EEG] activity) and subjective (NASA-TLX) measurements. EEG theta power and NASA-TLX (mental workload) scores were significantly higher for high than low difficulty levels. However, both EEG and NASA-TLX responses were reduced with repeated training in the second session. These findings highlight the effectiveness of EEG in continuous monitoring of workload variation caused by task difficulty and implementing training programs to moderate cognitive workload for forklift operators.
  • Thumbnail Image
    A data-driven approach to understand factors contributing to exoskeleton use-intention in construction
    Kim, Sunwook; Moore, Albert; Ojelade, Aanuoluwapo; Gutierrez, Nancy; Harris-Adamson, Carisa; Barr, Alan; Srinivasan, Divya; Nussbaum, Maury A. (SAGE, 2023-10-25)
    Work-related musculoskeletal disorders (WMSDs) remain an important heath concern for construction workers. Occupational exoskeletons (EXOs) are a new ergonomic intervention to control WMSD risk, yet their adoption has been low in construction. We explored contributing factors to EXO use-intention, by building a decision tree to predict the intention to try an exoskeleton using responses to an online survey. Variable selection and hyperparameter tuning were used respectively to reduce the number of potential predictors, and for a better prediction performance. Performance was assessed using four common metrics. The importance of variables in the final tree was calculated to understand which variable had a greater influence. The final tree had moderate prediction performance. Important variables identified were associated with opinions on EXO use, demographics, job demands, and perceived potential risks. The key influential variables were EXOs becoming standard equipment and fatigue reduction with EXO use. Practical implications of the findings are discussed.
  • Thumbnail Image
    Effects of an arm-support exoskeleton on perceived work intensity and musculoskeletal discomfort: An 18-month field study in automotive assembly
    Kim, Sunwook; Nussbaum, Maury A.; Smets, Marty; Ranganathan, Shyam (Wiley, 2021-08-06)
    Background: Exoskeleton (EXO) technologies are a promising ergonomic intervention to reduce the risk of work-related musculoskeletal disorders, with efficacy supported by laboratory- and field-based studies. However, there is a lack of field-based evidence on long-term effects of EXO use on physical demands. Methods: A longitudinal, controlled research design was used to examine the effects of arm-support exoskeleton (ASE) use on perceived physical demands during overhead work at nine automotive manufacturing facilities. Data were collected at five milestones (baseline and at 1, 6, 12, and 18 months) using questionnaires. Linear mixed models were used to understand the effects of ASE use on perceived work intensity and musculoskeletal discomfort (MSD). Analyses were based on a total of 41 participants in the EXO group and 83 in a control group. Results: Across facilities, perceived work intensity and MSD scores did not differ significantly between the EXO and control groups. In some facilities, however, neck and shoulder MSD scores in the EXO group decreased over time. Wrist MSD scores in the EXO group in some facilities remained unchanged, while those scores increased in the control group over time. Upper arm and low back MSD scores were comparable between the experimental groups. Conclusion: Longitudinal effects of ASE use on perceived physical demands were not found, though some suggestive results were evident. This lack of consistent findings is discussed, particularly supporting the need for systematic and evidence-based ASE implementation approaches in the field that can guide the optimal selection of a job for ASE use.
  • Thumbnail Image
    Effects of back-support exoskeleton use on gait performance and stability during level walking
    Park, Jang-Ho; Kim, Sunwook; Nussbaum, Maury A.; Srinivasan, Divya (Elsevier, 2022-02-01)
    Background: Back-support exoskeletons (BSEs) are a promising intervention to mitigate physical demands at work. Although growing evidence indicates that BSEs can reduce low-back physical demands, there is limited understanding of potential unintended consequences of BSE use, including the risk of falls. Research question: Does using a BSE adversely affect gait performance and stability, and are such effects dependent on specific BSE external torque characteristics? Methods: Twenty participants (10 M, 10 F) completed five level over-ground walking trials and a five-minute treadmill walking trial while wearing a BSE (backX™) with three different levels of external torque (i.e., no torque, low torque, and high torque) and in a control (no-exoskeleton) condition. Spatiotemporal gait patterns, stride-to-stride gait variability measures, required coefficient-of-friction (RCoF), and minimum foot clearance (MFC) were determined, to assess gait performance. Gait stability was quantified using the maximum Lyapunov exponent (MLE) of trunk kinematics and the margin-of-stability (MoS). Results: Using the backX™ with high supportive torque decreased slip risk (7% decrease in RCoF) and slightly improved trunk stability (3% decrease in MLE). However, it also decreased step length (1%), increased step width (10%) and increased gait variability (8–19%). Changes in MoS were complex: while MoS at heel strike decreased in the AP direction, it increased in the ML direction. There was a rather large decrease in MoS (26%) in the ML direction during the swing phase. Significance: This is the first study to quantify the effects of wearing a passive BSE with multiple supportive torque levels on gait performance and stability during level walking. Our results, showing that the external torque of the BSE may adversely affect gait step width, variability, and dynamic stability, can contribute to better design and practice guidelines to facilitate the safe adoption of BSEs in the workplace.
  • Thumbnail Image
    Effects of Back-support Exoskeletons on Task Performance and Perceived Exertion During Construction-related Tasks: Differences related to Experience
    Ojelade, Aanuoluwapo; Kim, Sunwook; Morris, Wallace; Harris-Adamson, Carisa; Barr, Alan; Nussbaum, Maury A. (SAGE, 2023-10-24)
    The effects of three Back-support Exoskeletons (BSEs) on task performance (completion time) and perceived exertion were examined during simulated construction tasks. Forty participants (20 novices) completed the tasks using no BSE and three BSEs, each with two support settings (On and Off). There were significant main and interaction effects of Experience level on both outcome measures. Using BSEs negatively affected task performance for some tasks, though these effects were smaller among experienced participants. Interestingly, using BSEs inconsistently affected perceived physical effort, possibly due to the characteristics of the tasks, BSE designs, and experience levels. These results indicate that the effects of BSE on task performance and perceived physical effort are specific to experience level, BSE designs, and task demands. Further study is needed to better understand how to generalize these results to the actual work population.
  • Thumbnail Image
    Effects of Back-support Exoskeletons on Task Performance and Usability During Simulated Construction-relevant Tasks
    Ojelade, Aanuoluwapo; Morris, Wallace; Kim, Sunwook; Harris-Adamson, Carisa; Barr, Alan; Nussbaum, Maury A. (SAGE, 2022-10-11)
  • Thumbnail Image
    Effects of using a whole-body powered exoskeleton during simulated occupational load-handling tasks: A pilot study
    Park, Hanjun; Kim, Sunwook; Nussbaum, Maury A.; Srinivasan, Divya (Elsevier, 2022-01-01)
    Whole-body powered exoskeletons (WB-PEXOs) can be effective in reducing the physical demands of heavy occupational work, yet almost no empirical evidence exists on the effects of WB-PEXO use. This study assessed the effects of WB-PEXO use on back and leg muscle activities during lab-based simulations of load handling tasks. Six participants (4M, 2F) completed two such tasks (load carriage and stationary load transfer), both with and without a WB-PEXO, and with a range of load masses in each task. WB-PEXO use reduced median levels of muscle activity in the back (∼42–53% in thoracic and ∼24–43% in lumbar regions) and legs (∼41–63% in knee flexors and extensors), and mainly when handling loads beyond low-moderate levels (10–15 kg). Overall, using the WB-PEXO also reduced inter-individual variance (smaller SD) in muscle activities. Future work should examine diverse users, focus on finding effective matches between WB-PEXO use and specific tasks, and identify applications in varied work environments.
  • Thumbnail Image
    Effects of using passive back- and arm-support exoskeletons for cart pushing and pulling
    Park, Hanjun; Noll, Alex; Kim, Sunwook; Nussbaum, Maury A. (SAGE, 2023-10-20)
    Though studies have shown both BSEs and ASEs are effective in reducing physical demands for various manual material handling tasks, limited evidence is currently available on their effects in pushing and pulling tasks. We evaluated the impacts of using a passive back-support (BSE) and an arm-support exoskeleton (ASE) on trunk kinematics, and the back and arm musculature for pushing and pulling of a moderately loaded (100 kg) cart. Fourteen volunteers performed cart the tasks, each of whom was randomly assigned one of the exoskeletons (BSE or ASE). Wearing the BSE substantially reduced lumbar muscle activity during both pushing (up to ~39%) and pulling (up to ~35%) compared to not wearing EXOs, while wearing an ASE had no significant impacts. For the anterior deltoid muscle, neither BSE nor ASE had beneficial impacts. Findings from the current study help to understand the effects of BSEs and ASEs in pushing and pulling tasks.
  • Thumbnail Image
    Evaluation of Two Approaches for Aligning Data Obtained from a Motion Capture System and an In-Shoe Pressure Measurement System
    Kim, Sunwook; Nussbaum, Maury A. (MDPI, 2014-09-12)
    An in-shoe pressure measurement (IPM) system can be used to measure center of pressure (COP) locations, and has fewer restrictions compared to the more conventional approach using a force platform. The insole of an IPM system, however, has its own coordinate system. To use an IPM system along with a motion capture system, there is thus a need to align the coordinate systems of the two measurement systems. To address this need, the current study examined two different approaches—rigid body transformation and nonlinear mapping (i.e., multilayer feed-forward neural network (MFNN))—to express COP measurements from an IPM system in the coordinate system of a motion capture system. Ten participants (five male and five female) completed several simulated manual material handling (MMH) activities, and during these activities the performance of the two approaches was assessed. Results indicated that: (1) performance varied between MMH activity types; and (2) a MFNN performed better than or comparable to the rigid body transformation, depending on the specific input variable sets used. Further, based on the results obtained, it was argued that a nonlinear mapping vs. rigid body transformation approach may be more effective to account for shoe deformation during MMH or potentially other types of physical activity.
  • Thumbnail Image
    An exploratory study comparing three work/rest schedules during simulated repetitive precision work
    Tsao, Liuxing; Kim, Sunwook; Ma, Liang; Nussbaum, Maury A. (Taylor & Francis, 2021-08-03)
    The pattern of work and rest can influence both physical fatigue and task performance in manual operations. However, there is relatively limited evidence regarding the influences of specific work/rest schedules in tasks requiring high repetitiveness and precision demands, along with relatively low exertion levels. Eighteen participants completed an exploratory study that simulated such tasks, to compare the effects of three distinct work/rest schedules (i.e. short frequent [short] and long infrequent breaks [long], and a self-selected schedule) on muscle fatigue, task performance (in terms of accuracy and speed), and preference. Schedules with long or self-selected breaks generally induced less muscle fatigue, compared with the short break condition. Participants preferred the self-selected condition the most and the long-break condition the least. The different schedules tested did not influence task performance. A self-selected schedule may be beneficial for repetitive precision task, to achieve a balance across muscle fatigue, task performance, and individual preference. Practitioner summary: Influences of three work/rest schedules (i.e. short and long breaks, and a self-selected schedule) on fatigue, performance, and preference were explored during repetitive precision tasks. Schedules with long or self-selected breaks induced less muscle fatigue and none of the three schedules influenced performance. A self-selected schedule was the most preferred.
  • Thumbnail Image
    Forklift Driving Performance of Novices with Repeated VR-based Training
    Islam, Md Shafiqul; Jamshid Nezhad Zahabi, Saman; Kim, Sunwook; Lau, Nathan; Nussbaum, Maury A.; Lim, Sol (SAGE, 2023-10-19)
    Virtual reality (VR) has emerged as a promising tool for training novice forklift drivers, but temporal patterns of such improvements are largely unknown. We trained 19 novice participants using an order-picker VR simulator on a selected driving lesson. In two sessions, participant driving performance was assessed using task completion time and kinematics of the head, shoulder, and lumbar spine via inertial measurement units (IMUs). Completion time and head flexion/movement decreased significantly (up to 22.4% and 31.5%, respectively). The observed changes in head motion (flexion/extension) indicate an initial adjustment period to prepare a mental model of the driving task and the control panel, which was also adapted over repeated trials. One implication of our results is that reduced head flexion/extension could be used as an indication of a novice driver’s improved skill during the early stages of training, in terms of familiarizing themselves with vehicle control and the vehicle control panel.
  • Thumbnail Image
    Gait kinematics when learning to use a whole-body powered exoskeleton
    Park, Hanjun; Lee, Youngjae; Kim, Sunwook; Nussbaum, Maury A.; Srinivasan, Divya (SAGE, 2021-09)
  • Thumbnail Image
    Human Gait During Level Walking With an Occupational Whole-Body Powered Exoskeleton: Not Yet a Walk in the Park
    Kim, Sunwook; Srinivasan, Divya; Nussbaum, Maury A.; Leonessa, Alexander (IEEE, 2021-04-01)
    With rapid advancements in exoskeleton technologies, a whole-body powered exoskeleton (WB-PEXO) for augmenting human physical capacity (a ``super-operator'') is generating increasing attention as an integral part of Industry 4.0. Our understanding of WB-PEXO use is lagging, however, largely due to the lack of detailed evaluations via human-subjects testing of a WB-PEXO.We examined (independently from the manufacturer of a WB-PEXO) the potential impacts of using a state-of-the-art WB-PEXO prototype (pre-alpha prototype version of the Sarcos Guardian® XO®) on users (n = 5) during a common basic activity in the workplace, level walking. With emphasis on the ``human'', impacts of XO use (compared to a no EXO baseline) were assessed in terms of lower limb intersegmental coordination, muscle activity, and postural dynamic stability. A larger variance between participants was observed for intersegmental coordination with XO use, and participants appeared to rely on more hip motions. When using the XO, participants exhibited higher muscle activity levels in the lower limb muscle groups monitored. Further, there was a moderate to high similarity in muscle activity profiles between the XO and no EXO conditions (RXY (T) = 0:70 - 0:92), yet muscle activity profiles when using the XO were generally time-lagged from those without the XO. We discuss the results within the context of developing a mental model for walking with the XO, and aspects of human-robot interaction such as transparency of the XO and understanding user state and intention. In concluding, we outline several future research topics for occupational WB-PEXO development.
  • Thumbnail Image
    An indoor fall monitoring system: Robust, multistatic radar sensing and explainable, feature-resonated deep neural network
    Shen, Menqi; Tsui, Kwok-Leung; Nussbaum, Maury A.; Kim, Sunwook; Lure, Fleming (IEEE, 2023-04)
    Indoor fall monitoring is challenging for community-dwelling older adults due to the need for high accuracy and privacy concerns. Doppler radar is promising, given its low cost and contactless sensing mechanism. However, the line-of-sight restriction limits the application of radar sensing in practice, as the Doppler signature will vary when the sensing angle changes, and signal strength will be substantially degraded with large aspect angles. Additionally, the similarity of the Doppler signatures among different fall types makes it extremely challenging for classification. To address these problems, in this paper we first present a comprehensive experimental study to obtain Doppler radar signals under large and arbitrary aspect angles for diverse types of simulated falls and daily living activities. We then develop a novel, explainable, multi-stream, feature-resonated neural network (eMSFRNet) that achieves fall detection and a pioneering study of classifying seven fall types. eMSFRNet is robust to both radar sensing angles and subjects. It is also the first method that can resonate and enhance feature information from noisy/weak Doppler signatures. The multiple feature extractors - including partial pre-trained layers from ResNet, DenseNet, and VGGNet - extracts diverse feature information with various spatial abstractions from a pair of Doppler signals. The feature-resonated-fusion design translates the multi-stream features to a single salient feature that is critical to fall detection and classification. eMSFRNet achieved 99.3% accuracy detecting falls and 76.8% accuracy for classifying seven fall types. Our work is the first effective multistatic robust sensing system that overcomes the challenges associated with Doppler signatures under large and arbitrary aspect angles, via our comprehensible feature-resonated deep neural network. Our work also demonstrates the potential to accommodate different radar monitoring tasks that demand precise and robust sensing.
  • Thumbnail Image
    Multigrid Accelerated Cellular Automata for Structural Optimization: A 1-D Implementation
    Kim, Sunwook (Virginia Tech, 2002-12-16)
    Multigrid acceleration is typically used for the iterative solution of partial differential equations in physics and engineering. A typical multigrid implementation uses a base discretization method, such as finite elements or finite differences, and a set of successively coarser grids that is used for accelerating the convergence of the iterative solution on the base grid. The presented thesis extends the use of multigrid acceleration to the design optimization of a sample structural system and demonstrates it within the context of the recently introduced Cellular Automata paradigm for design optimization. Within the design context, the multigrid scheme is not only used for accelerating the analysis iterations, but is also used to help refine the design across multiple grid levels to accelerate the design convergence. A comparison of computational efficiencies achieved by different multigrid implementations, including the multigrid accelerated nested design iteration scheme, is presented. The method is described in its generic form which can be applicable not only to the Cellular Automata paradigm but also to more general finite element analysis based design schemes as well.
  • Thumbnail Image
    Muscle Activation Patterns and Chronic Neck-Shoulder Pain in Computer Work
    Kelson, Denean M. (Virginia Tech, 2018-04-20)
    Prolonged computer work is associated with high rates of neck and shoulder pain symptoms, and as computers have become increasingly more common, it is becoming critical that we develop sustainable interventions targeting this issue. Static muscle contractions for prolonged periods often occur in the neck/shoulder during computer work and may underlie muscle pain development in spite of rather low relative muscle load levels. Causal mechanisms may include a stereotypical recruitment of low threshold motor units (activating type I muscle fibers), characterized by a lack of temporal as well as spatial variation in motor unit recruitment. Based on this theory, although studies have postulated that individuals with chronic neck-shoulder pain will show less variation in muscle activity compared to healthy individuals when engaged in repetitive/monotonous work, this has seldom been verified in empirical studies of actual computer work. Studies have rarely addressed temporal patterns in muscle activation, even though there is a consensus that temporal activation patterns are important for understanding fatigue and maybe even risks of subsequent musculoskeletal disorders. This study applied exposure variation analysis (EVA) to study differences in temporal patterns of trapezius muscle activity as individuals with and without pain performed computer work. The aims of this study were to: Assess the reliability of EVA to measure variation in trapezius muscle activity in healthy individuals during the performance of computer work; Determine the extent to which healthy subjects differ from those with chronic pain in trapezius muscle activity patterns during computer work, measured using EVA. Thirteen touch-typing, right-handed participants were recruited in this study (8 healthy; 5 chronic pain). The participants were asked to complete three 10-minute computer tasks (TYPE, CLICK and FORM) in two pacing conditions (self-paced, control-paced), with the healthy group completing two sessions and the pain group completing one. Activation of the upper trapezius muscle was measured using surface electromyography (EMG). EMG data were organized into 5x5 EVA matrices with five amplitude classes (0-6.67, 6.67-20, 20-46.67, 46.67-100, >100% Reference Voluntary Exertion) and five duration classes (0- 1, 1-3, 3-7, 7-15, >15 seconds). EVA marginal distributions (along both amplitude and duration classes) for each EVA class, as well as summary measures (mean and SD) of the marginal sums along each axis were computed. Finally, “resultant” mean and SD across all EVA cells were computed. The reliability in EVA indices was estimated using intra-class correlation coefficients (ICC), coefficient of variation (CV) and standard error of measurement (SEM), computed from repeated measurements of healthy individuals (aim 1), and EVA indices were compared between groups (aim 2). Reliability of EVA amplitude marginal sums ranged from moderate to high in the self-paced condition and low to moderate in the control-paced condition. The duration marginal sums were moderate in the self-paced condition and moderate to high in the control-paced condition. The summary measures (means and SDs) were moderate to high in both the self-paced and control-paced condition. Group comparisons revealed that individuals with chronic pain spent longer durations of work time in higher EVA duration categories, exhibited larger means along the amplitude, duration and in the resultant, and higher EVA SD in the amplitude and duration axes as compared to the healthy group. To our knowledge, this is the first study to report on the reliability of EVA applied specifically to computer work. Furthermore, EVA was used to assess differences in muscle activation patterns as individuals with and without chronic pain engaged in computer work. Individuals in the pain group seemed to exhibit prolonged sustained activation of the trapezius muscle to a significantly greater extent than controls, even though they did not experience pain during the performance of the computer tasks (as obtained through self-reports). Thus, these altered muscle recruitment patterns observed in the pain subjects, even in the absence of task-based pain/discomfort, are suggestive of chronic motor control changes occurring in adaptation to pain, and may have implications for the etiology of neck and upper-limb musculoskeletal disorders.
  • Thumbnail Image
    A novel approach to quantify the assistive torque profiles generated by passive back-support exoskeletons
    Madinei, Saman; Kim, Sunwook; Park, Jang-Ho; Srinivasan, Divya; Nussbaum, Maury A. (Elsevier, 2022-10-31)
    Industrial exoskeletons are a promising ergonomic intervention to reduce the risk of work-related musculoskeletal disorders by providing external physical support to workers. Passive exoskeletons, having no power supplies, are of particular interest given their predominance in the commercial market. Understanding the mechanical behavior of the torque generation mechanisms embedded in passive exoskeletons is, however, essential to determine the efficacy of these devices in reducing physical loads (e.g., in manual material handling tasks). We introduce a novel approach using a computerized dynamometer to quantify the assistive torque profiles of two passive back-support exoskeletons (BSEs) at different support settings and in both static and dynamic conditions. The feasibility of this approach was examined using both human subjects and a mannequin. Clear differences in assistive torque magnitudes were evident between the two BSEs, and both devices generated more assistive torques during trunk/hip flexion than extension. Assistive torques obtained from human subjects were often within similar ranges as those from the mannequin, though values were more comparable over a narrow range of flexion/extension angles due to practical limitations with the dynamometer and human subjects. Characterizing exoskeleton assistive torque profiles can help in better understanding how to select a torque profile for given task requirements and user anthropometry, and aid in predicting the potential impacts of exoskeleton use by incorporating measured torque profiles in a musculoskeletal modeling system. Future work is recommended to assess this approach for other occupational exoskeletons.