Conference PaperPDF Available

EMG analysis of an upper body exoskeleton during automotive assembly

Authors:
Jason C Gillette at Iowa State University
Jason C Gillette
Mitchell L. Stephenson at University of Montana Western
Mitchell L. Stephenson
Download full-text PDF
Read full-text
Download citation
Content uploaded by Jason C Gillette
Author content
All content in this area was uploaded by Jason C Gillette on Aug 23, 2018
Content may be subject to copyright.
42nd Annual Meeting of the American Society of Biomechanics, Rochester, MN, USA, August 8th – 11th, 2018
EMG ANALYSIS OF AN UPPER BODY EXOSKELETON DURING AUTOMOTIVE ASSEMBLY
Jason C. Gillette and Mitchell L. Stephenson
Iowa State University, Ames, IA, USA
email: gillette@iastate.edu
INTRODUCTION
Shoulder injuries in the workplace result in the
longest recovery time of any body part: 23 median
days missed [1]. Common causes of shoulder pain
include bursitis, tendinitis, impingement, instability,
and arthritis [2]. Repetitive overhead postures are
widely thought to increase the risk of shoulder injury.
Threshold limit values (TLV) for upper limb fatigue
have been developed that relate maximum voluntary
contraction (MVC) to duty cycle [3]. Using this
equation, a 39.5% MVC should not be exceeded with
a 10% duty cycle, and a 16.5% MVC should not be
exceeded with a 50% duty cycle.
When worksite modifications are impractical or
impossible, robotic or passive assistive devices may
be utilized in an attempt to prevent shoulder injuries.
The Levitate Airframe is a passive exoskeleton
designed to support arm weight during overhead
shoulder postures. Previous testing with workers at
John Deere indicated promising reductions of deltoid
muscle activity when using this exoskeleton [4]. The
purpose of this study was to assess this exoskeleton
during overhead automotive assembly at Toyota
Canada. Our hypothesis was that deltoid muscle
activity would be reduced with the exoskeleton.
METHODS
Eleven experienced male workers (age 357 yr)
volunteered for this study. Data were collected for
ten overhead automotive assembly tasks. One
participant performed three job tasks, five
participants performed two job tasks, and five
participants performed one job task. Eight job tasks
were performed by two participants, and two job
tasks were performed by one participant (18
comparisons). Approximately 12 minutes of data
were collected per job task, with nine job tasks
having ten repetitions (automotives completed), and
one job task having three repetitions (multiple
stations). Data for each job task were collected with
and without use of the exoskeleton.
IRB-approved informed consent was obtained from
all participants. Data were collected using a Delsys
Trigno wireless electromyography (EMG) system at
a sampling frequency of 1926 Hz using Delsys
EMGworks software. EMG sensors were placed on
eight muscles: right and left anterior deltoid, biceps
brachii, superior trapezius, and lumbar erector
spinae. Maximum voluntary isometric contractions
(MVIC) were performed in the following postures:
shoulder abduction empty can, seated elbow flexion,
and prone spinal extension. EMG data were analyzed
using custom written code in Matlab.
EMG signals were inspected to remove any non-
physiological artifacts. The EMG data were
bandpass filtered with a 4th-order zero-lag
Butterworth filter from 20-450 Hz and then rectified.
Next, the EMG data were low-pass filtered at 10 Hz
to create a linear envelope. A moving window was
used to determine the maximum one second of
sustained EMG amplitudes for the MVICs. The job
task EMG amplitudes were divided into consecutive
one second intervals. The highest 10% and 50% of
EMG amplitudes were determined as a measure of
10% and 50% duty cycles.
Maximum 10% and 50% EMG amplitudes were
normalized by the maximum one second MVICs
(%MVIC). EMG amplitudes presented here are for
the dominant arm anterior deltoid, biceps brachii,
and trapezius muscles. The right and left side erector
spinae muscle results were averaged. Paired t-tests
were used to compare EMG amplitudes between
conditions (with and without the exoskeleton) in all
job tasks, with significance indicated at p < 0.03
(adjusted for 18 comparisons with 11 participants).
Maximum 50% deltoid amplitudes are presented for
individual job tasks (Figure 1) to explore if some
tasks benefited differently from exoskeleton usage.
42nd Annual Meeting of the American Society of Biomechanics, Rochester, MN, USA, August 8th – 11th, 2018
RESULTS AND DISCUSSION
Maximum 10% EMG amplitudes were significantly
reduced for the deltoid (p=0.02), biceps brachii
(p=0.01), and erector spinae (p=0.01) when wearing
the exoskeleton (Table 1). There was not a significant
change in trapezius EMG amplitudes (p=0.35). The
deltoid demonstrated the greatest magnitude of EMG
amplitude change: a 6.5% MVIC reduction with the
exoskeleton. Deltoid EMG standard deviations
exceeded the 39.5% TLV without the exoskeleton
but fell below this level with the exoskeleton.
Trapezius EMG standard deviations exceeded the
39.5% TLV both without and with the exoskeleton.
Maximum 50% EMG amplitudes were significantly
reduced for the deltoid (p<0.01), biceps brachii
(p<0.01), and erector spinae (p=0.03) when wearing
the exoskeleton (Table 1). There was not a significant
change in trapezius EMG amplitudes (p=0.61). The
deltoid showed the greatest magnitude of EMG
amplitude change: a 4.4% MVIC reduction with the
exoskeleton. Deltoid EMG average exceeded the
16.5% TLV without the exoskeleton, but the average
and standard deviation fell below this level with the
exoskeleton. Trapezius EMG averages exceeded the
16.5% TLV both without and with the exoskeleton.
Figure 1: Reduction in maximum 50% EMG
amplitudes for the deltoid as a function of job task.
Our hypothesis was supported by the significant
reduction in deltoid EMG amplitudes when wearing
the exoskeleton. Thus, the exoskeleton appeared to
support the primary deltoid functions of shoulder
flexion and abduction. Reductions in biceps brachii
EMG may be due to exoskeleton stabilization of the
shoulder to indirectly support elbow flexion. The
unexpected reductions in erector spinae EMG may
be explained by the exoskeleton reducing anterior
lean and promoting a more upright posture. Effects
on the trapezius were mixed, with possible benefits
of shoulder abduction support mixed with possible
shoulder strap restriction of scapular elevation.
We suggest assessing specific job tasks to determine
potential benefits of assistive devices. The Levitate
Airframe is most likely to benefit repetitive overhead
movements that challenge the deltoid. When
considering the different job tasks, there were tasks
where the exoskeleton was more beneficial than
others (Figure 1). We suggest that it is important to
consider EMG amplitudes and standard deviations
relative to TLVs as an indicator of fatigue and injury
risk. In addition, analyzing maximum 10% and 50%
EMG amplitudes may be useful for assessing risks
associated with low vs. high repetition tasks.
REFERENCES
1. BLS. Nonfatal occupational injuries and illnesses
requiring days away from work. 2016.
2. AAOS. OrthoInfo. 2010.
3. ACGIH. Threshold Limit Values (TLVs). 2016.
4. Gillette and Stephenson. ASB Proceedings.
Boulder, CO, USA, 2017.
ACKNOWLEDGEMENTS
The authors would like to acknowledge Terry Butler
(Lean Steps Consulting), Seth Burt (Toyota), Alex
Mason (Toyota), and Joseph Zawaideh (Levitate) for
their assistance in coordinating this study.
Table 1: Maximum 10% and 50% EMG amplitudes with and without the exoskeleton
Maximum 10% EMG Amplitude
Maximum 50% EMG Amplitude
(%MVIC)
Deltoid
Biceps
Trap
Spinae
Deltoid
Trap
Spinae
Without
31.8±8.2
14.9±4.2
28.0±12.9
26.0±9.5
16.6±6.9
17.8±9.3
16.8±6.3
Exoskeleton
25.4±6.4*
13.5±3.9*
30.8±13.9
21.9±6.6*
12.2±3.3*
18.7±9.0
14.5±4.4*
*indicates significantly reduced EMG amplitude when wearing the exoskeleton (p < 0.03)
0
2
4
6
8
10
12
Job
1Job
2Job
3Job
4Job
5Job
6Job
7Job
8Job
9Job
10
EMG Reduction (%MVIC)
... In recent studies on (upper-body) exoskeleton use, researchers have already explored various aspects related to muscle activity, but with varying results (5)(6)(7)(8)(9)(10). The need to support the back and shoulder muscles is particularly important for overhead and lifting activities (5)(6)(7)(8)(9)(10)(11). ...
... In recent studies on (upper-body) exoskeleton use, researchers have already explored various aspects related to muscle activity, but with varying results (5)(6)(7)(8)(9)(10). The need to support the back and shoulder muscles is particularly important for overhead and lifting activities (5)(6)(7)(8)(9)(10)(11). Studies have shown a reduction in muscular activation associated with exoskeleton use in the M. deltoideus, M. Biceps brachii during overhead tasks, and M. Triceps brachii (7, 8; 11). ...
... In addition, limited evidence exists regarding gender differences in muscle activity during exoskeleton use, with some studies indicating higher muscular demands and discomfort for females during overhead and carrying tasks (6,32,33). Moreover, to our knowledge, no additional research has specifically examined gender disparities in the use of upper body exoskeletons for tasks involving overhead movements or carrying/lifting activities. ...
Gender differences in the use of an upper-extremity exoskeleton during physically and cognitively demanding tasks- a study protocol for a randomized experimental trial
Article
Full-text available
  • Oct 2024
Background Upper limb exoskeletons are recommended to alleviate muscle fatigue, particularly in working conditions inducing musculoskeletal discomfort like overhead work. However, wearing an exoskeleton might introduce cognitive-motor interference, affecting performance. Understanding its neural impact and potential gender differences in design effects is crucial. Therefore, the aim of this study is to examine exoskeleton effects addressing cross-gender comparisons, and exploring the impact on cognitive and physical workload in real-world scenarios. The research questions address the impact of exoskeleton use on muscle synergies, upper body posture, cognitive resources, comfort/discomfort, acceptance and usability. Methods The cross-sectional study integrates a multifactorial mixed-measure design. Participants are grouped by gender (female vs. male) and working condition (with vs. without exoskeleton). Motor performance and underlying neuronal correlates (fNIRS) will be analyzed. Based on an a priori sample size calculation, 80 participants (40 female/40 male) will be recruited. Working performance will be assessed by 1. Physical Performance Task (PILE task) and 2. Precision Task (following the Fitts paradigm), while body postures will be monitored with an Xsens motion capture system. Brain activation will be captured with an fNIRS system comprising 32 active optodes. Postural comfort/discomfort, acceptance, and usability will be reported via standardized questionnaires. Discussion The study will gain insights into potential gender differences in exoskeleton use and will contribute to designing and optimizing the implementation of exoskeletons by considering muscle synergies, movement variability and cognitive resource allocation. Additionally, the study also highlights user discomfort, a crucial factor that could impede widespread adoption, particularly among females, in real-world scenarios.
View
... Thus, passive exoskeletons may be more easily used in industry to reduce MSDs that widely affect workers. Indeed, according to the literature, the use of passive exoskeletons decreases the EMG amplitude of the arm and shoulder muscles, notably the anterior deltoid, medialis deltoid, upper trapezius and biceps brachii [3,[14][15][16][17][18][19][20][21][22][23][24][25][26][27][28]). There is also a subsequent decrease in fatigue when the subject wears the exoskeleton compared to situations without the exoskeleton [29][30][31][32]. ...
... According to some studies, the use of exoskeletons could be accompanied by an increase in the muscular activity of other muscles, in particular lumbar erector spinae and triceps brachii [14,16,22,27]. However, other investigations did not highlight this increase in EMG amplitude for the triceps brachii [3] and lumbar erector spinae [18,20]. These adverse outcomes may be related to the fact that studies have been carried out either into a laboratory or directly in the field. ...
... These adverse outcomes may be related to the fact that studies have been carried out either into a laboratory or directly in the field. Of all the studies with passive upper limb exoskeletons, only nine have been conducted in the field [19,20,23,26,29,[33][34][35][36]. Results of these studies are mostly in agreement with those carried out into a laboratory environment, with a fatigue reduction and a decrease of muscle solicitation for shoulder muscles, while wearing the exoskeleton. ...
Evaluation of a Passive Upper Limb Exoskeleton in Healthcare Workers during a Surgical Instrument Cleaning Task
Article
Full-text available
(1) Background: Healthcare workers are highly affected by work-related musculoskeletal disorders, particularly in the lower back, neck and shoulders, as their occupational tasks expose them to biomechanical constraints. One solution to prevent these musculoskeletal disorders may be the use of a passive exoskeleton as it aims to reduce muscle solicitation. However, few studies have been carried out directly in this field to assess the impact of the use of a passive upper limb exoskeleton on this population. (2) Methods: Seven healthcare workers, equipped with electromyographic sensors, performed a tool cleaning task with and without a passive upper limb exoskeleton (Hapo MS, Ergosanté Technologie, France). Six muscles of the upper limbs were analysed, i.e., anterior deltoid, biceps brachii, pectoralis major, latissimus dorsi, triceps brachii and longissimus thoracis. A subjective analysis of the usability of the equipment, the perception of effort and discomfort, was also carried out using the System Usability Scale and the Borg scale. (3) Results: The longissimus thoracis was the most used muscle during this task. We observed a significant decrease in the muscular solicitation of the anterior deltoid and latissimus dorsi when wearing the exoskeleton. Other muscles were not significantly impacted by the device. (4) Conclusions: the passive exoskeleton used in this study allowed the reduction in muscular load on the anterior deltoid and latissimus dorsi without negative effects on other muscles. Other field studies with exoskeletons are now necessary, particularly in hospitals, to increase our knowledge and improve the acceptability of this system for the prevention of musculoskeletal disorders.
View
... Hence, the Automotive industry has the highest number of publications, corresponding to 35.56%. Within the Automotive studies, two were TRL-5 [48,56] and two were TRL-6 [57,58], eleven were TRL-7 [49,[59][60][61][62][63][64][65][66][67][68], and one included both TRL-6 and 7 [28]. Ship construction has only one study at TRL-7 [69], while Health and social care includes two papers at TRL-5 [70,71] and one at TRL-6 and 7 [72]. ...
... At TRL-6, EMG was solely used in 4 papers, with a further 3 papers reporting EMG along with other assessments (7 of 8 papers -87%) [28,55,57,58,76,85,86]. Finally, at TRL-7, EMG was solely reported in 8 cases with 4 papers using some combination of assessment formats (12 of 31 papers -38%) [28,49,54,55,60,61,63,64,67,73,82,83]. ...
View
... Although many studies have been reported in the last few years (Google Scholar produced 1340 references on 28 November 2023 for the keywords Exoskeleton, EMG, and Workplace), most of them were related to the comparison of individual muscle activation with and without an exoskeleton [41][42][43]. ...
Assessment of Muscle Coordination Changes Caused by the Use of an Occupational Passive Lumbar Exoskeleton in Laboratory Conditions
Article
Full-text available
  • Dec 2023
  • SENSORS-BASEL
The introduction of exoskeletons in industry has focused on improving worker safety. Exoskeletons have the objective of decreasing the risk of injury or fatigue when performing physically demanding tasks. Exoskeletons’ effect on the muscles is one of the most common focuses of their assessment. The present study aimed to analyze the muscle interactions generated during load-handling tasks in laboratory conditions with and without a passive lumbar exoskeleton. The electromyographic data of the muscles involved in the task were recorded from twelve participants performing load-handling tasks. The correlation coefficient, coherence coefficient, mutual information, and multivariate sample entropy were calculated to determine if there were significant differences in muscle interactions between the two test conditions. The results showed that muscle coordination was affected by the use of the exoskeleton. In some cases, the exoskeleton prevented changes in muscle coordination throughout the execution of the task, suggesting a more stable strategy. Additionally, according to the directed Granger causality, a trend of increasing bottom-up activation was found throughout the task when the participant was not using the exoskeleton. Among the different variables analyzed for coordination, the most sensitive to changes was the multivariate sample entropy.
View
... [12] reported that studies on the stress applied to other joints were necessary, although lowerlimb devices effectively reduced lower-limb muscle activity. As such, wearing lower-and upper-limb exoskeleton devices can have both positive [13][14][15] and negative [16][17][18][19] effects, and, in particular, the effect of wearing an exoskeleton device may vary depending on the working environment, so further research is needed. ...
Evaluation of the Effects of Passive Lower-Limb Exoskeletons on Muscle Activities According to Working Heights
Article
Full-text available
  • Oct 2023
Introduction: Research and development efforts regarding passive lower-limb exoskeletons are actively ongoing to reduce work-related musculoskeletal disorders (WMDSs) from improper and prolonged posture. The aims of this study were to evaluate the effects of passive lower-limb exoskeletons and working heights on muscle activity and subjective comfort ratings. Methods: In this study, 20 males performed a 10-min drilling task for three levels of working height (60, 85, and 110 cm), and three levels of intervention (WO: without exoskeleton, WCEX, and WCC) were used as independent variables. The EMG data of eight muscles and subjective discomfort ratings in each of the six body parts were analyzed in this study. Results: The results of this study confirm that the effect of wearing a lower-limb exoskeleton device may vary depending on the muscle type and working height. Overall, the positive effects of wearing were generally observed in the lower-limb muscles, whereas the muscle activity of the upper limbs showed an increasing trend when wearing the device at a height of 85 cm or more. Conclusions: Therefore, to obtain positive effects in both the upper- and lower-limb muscles, using the lower-limb exoskeletons at a working height of 85 cm or less is recommended.
View
... Repetitive and prolonged overhead work is frequently observed in various industries including automobile and airplane manufacturing facilities (Rashedi et al., 2014;Maurice et al., 2019;Smets, 2019). Such overhead work is known to pose substantial musculoskeletal stress in the neck and upper extremities due to awkward postures and sustained loading, which are well-known risk factors associated with work-related musculoskeletal disorders (WMSDs) (Sluiter et al., 2001;Grieve and Dickerson, 2008;Rashedi et al., 2014;Gillette and Stephenson, 2018;Maurice et al., 2019). This in part explains the disproportionally high prevalence of WMSDs in manufacturing involving various overhead work (Chang et al., 2007). ...
Efficacy of passive upper-limb exoskeletons in reducing musculoskeletal load associated with overhead tasks
Article
  • Jan 2023
  • APPL ERGON
Overhead work can pose substantial musculoskeletal stress in many industrial settings. This study aimed to evaluate the efficacy of passive upper-limb exoskeletons in reducing muscular activity and subjective discomfort ratings. In a repeated-measures laboratory experiment, 20 healthy male participants performed 10-min drilling tasks with and without two passive upper-limb exoskeletons (VEX and Airframe). During the tasks, muscle activity in eight muscles (upper limb-upper trapezius, middle deltoid, biceps brachii, triceps brachii; low back-erector spinae; lower limb-rectus femoris, biceps femoris, tibialis anterior) was collected using electromyog-raphy as a physical exertion measure. Subjective discomfort rating in six body parts was measured using the Borg's CR-10 scale. The results showed that muscle activity (especially in the upper-limb muscles) was significantly decreased by 29.3-58.1% with both exoskeletons compared to no exoskeleton condition. The subjective discomfort ratings showed limited differences between the conditions. These findings indicate that passive upper-limb exoskeletons may have potential as an effective intervention to reduce muscular loading and physical exertion during overhead work.
View
... The purpose of this study was to determine if an upper body exoskeleton reduced muscle fatigue risk during automotive assembly job tasks at Toyota. Early results of this study have been previously reported (Gillette and Stephenson, 2018). This manuscript expands on that previous work by adding six additional team members, five additional assembly processes, and a revised ACGIH TLV analysis that includes the entire duty cycle curve (instead of just 10 and 50%). ...
Electromyography-based fatigue assessment of an upper body exoskeleton during automotive assembly
Article
Full-text available
  • Sep 2022
The purpose of this study was to assess an upper body exoskeleton during automotive assembly processes that involve elevated arm postures. Sixteen team members at Toyota Motor Manufacturing Canada were fitted with a Levitate Airframe, and each team member performed between one and three processes with and without the exoskeleton. A total of 16 assembly processes were studied. Electromyography (EMG) data were collected on the anterior deltoid, biceps brachii, upper trapezius, and erector spinae. Team members also completed a usability survey. The exoskeleton significantly reduced anterior deltoid mean active EMG amplitude (p = .01, Δ = −3.2 %MVC, d = 0.56 medium effect) and fatigue risk value (p < .01, Δ = −5.1 %MVC, d = 0.62 medium effect) across the assembly processes, with no significant changes for the other muscles tested. A subset of nine assembly processes with a greater amount of time spent in arm elevations at or above 90° (30 vs. 24%) and at or above 135° (18 vs. 9%) appeared to benefit more from exoskeleton usage. For these processes, the exoskeleton significantly reduced anterior deltoid mean active EMG amplitude (p < .01, Δ = −5.1 %MVC, d = 0.95 large effect) and fatigue risk value (p < .01, Δ = −7.4 %MVC, d = 0.96 large effect). Team members responded positively about comfort and fatigue benefits, although there were concerns about the exoskeleton hindering certain job duties. The results support quantitative testing to match exoskeleton usage with specific job tasks and surveying team members for perceived benefits/drawbacks.
View
View
Wearable upper limb robotics for pervasive health: A review
Article
Full-text available
  • Mar 2023
Wearable robotics, also called exoskeletons, have been engineered for human-centered assistance for decades. They provide assistive technologies for maintaining and improving patients’ natural capabilities towards self-independence and also enable new therapy solutions for rehabilitation towards pervasive health. Upper limb exoskeletons can significantly enhance human manipulation with environments, which is crucial to patients’ independence, self-esteem, and quality of life. For long-term use in both in-hospital and at-home settings, there are still needs for new technologies with high comfort, biocompatibility, and operability. The recent progress in soft robotics has initiated soft exoskeletons (also called exosuits), which are based on controllable and compliant materials and structures. Remarkable literature reviews have been performed for rigid exoskeletons ranging from robot design to different practical applications. Due to the emerging state, few have been focused on soft upper limb exoskeletons. This paper aims to provide a systematic review of the recent progress in wearable upper limb robotics including both rigid and soft exoskeletons with a focus on their designs and applications in various pervasive healthcare settings. The technical needs for wearable robots are carefully reviewed and the assistance and rehabilitation that can be enhanced by wearable robotics are particularly discussed. The knowledge from rigid wearable robots may provide practical experience and inspire new ideas for soft exoskeleton designs. We also discuss the challenges and opportunities of wearable assistive robotics for pervasive health.
View
Evaluation of a spring-loaded upper-limb exoskeleton in cleaning activities
Article
  • Jan 2023
  • APPL ERGON
In the past few years, companies have started considering the adoption of upper-limb occupational exoskeletons as a solution to reduce the health and cost issues associated with work-related shoulder overuse injuries. Most of the previous research studies have evaluated the efficacy of these devices in laboratories by measuring the reduction in muscle exertion resulting from device use in stereotyped tasks and controlled conditions. However, to date, uncertainties exist about generalizing laboratory results to more realistic conditions of use. The current study aims to investigate the in-field efficacy (through electromyography and perceived exertion), usability, and acceptance of a commercial spring-loaded upper-limb exoskeleton in cleaning job activities. The operators were required to maintain prolonged overhead postures while holding and moving a pole equipped with tools for window and ceiling cleaning. Compared to the normal working condition, the exoskeleton significantly reduced the total shoulder muscle activity (∼17%), the activity of the anterior deltoid (∼26%), medial deltoid (∼28%), and upper trapezius (∼24%). With the exoskeleton, the operators perceived reduced global effort (∼17%) as well as a reduced local effort in the shoulder (∼18%), arm (∼22%), upper back (∼14%), and lower back (∼16%). The beneficial effect of the exoskeleton and its suitability in cleaning settings are corroborated by the acceptance and usability scores assigned by operators, which averaged ∼5.5 out of 7 points. To the authors' knowledge, this study is the first to present an experience of exoskeleton use in cleaning contexts. The outcomes of this research invite further studies to test occupational exoskeletons in various realistic applications to foster scientific-grounded ergonomic evaluations and encourage the informed adoption of the technology.
View
Nonfatal occupational injuries and illnesses requiring days away from work
  • Jan 2016
BLS. Nonfatal occupational injuries and illnesses requiring days away from work. 2016.