Occupational exoskeletons: Insights from the lab to create a better working environment

Despite ongoing efforts to further improve the working conditions, 54% and 60% of the industrial workers in the European Union annually suffer from work-related musculoskeletal disorders at the shoulders and the back, respectively. In the European Union,this high injury burden costs up to C 422.1 billion every year. Work-related musculoskeletal disorders originate from the complex interplay of multiple factors, but me chanical load on the tissues is indisputably one of the fundamental contributors to the occurrence of these disorders. Reducing exposure to physical load may restore the balance between load-induced tissue damage and tissue repair. Such reduction may thus decrease the incidence of work-related musculoskeletal disorders and its concomitant costs.
Occupational exoskeletons are wearable assistive devices working in tandem with the user, where they act as an amplifier, aiming to restore, reinforce or augment hu man performance. These devices most frequently support the shoulder or the back, and have the potential to reduce the underlying factors associated with work-related musculoskeletal disorders, while preserving the highly valued flexibility and dexterity of the human worker. Most occupational exoskeletons are passive devices, where energy gets stored in elastic components. This energy is returned to assist the wearer. Companies re main hesitant to implement exoskeletons on large scale, since a comprehensive evidence base that documents the effect of these exoskeletons on the users is not yet available. A benchmarking framework directing researchers towards a more consistent assessment strategy could facilitate collection of such comprehensive evidence. Moreover, most exoskeleton assessments are currently conducted in the laboratory, but it remains un clear how the effect of an exoskeleton reported in a laboratory setting can be transferred to the field
The Exo4Work exoskeleton, a passive shoulder exoskeleton developed in an FWO SBO research project, aims to assist the shoulder joint during overhead work. Apart from that, this device aims to overcome three remaining challenges of current pas sive shoulder exoskeletons, i.e. kinematic compatibility, implementation of a metabolically convenient actuation system, and the amount of assistance a shoulder exoskeleton should provide. After construction of this exoskeleton, the effect of this new device on the user should be assessed during overhead work, but also tasks outside of the target zone of the exoskeleton should be included in the protocol.
The human workers are highly valued because of their flexibility and dexterity. Precision performance is one of those esteemed skills of the human operator, but no consensus has been reached regarding the effect of passive shoulder exoskeletons on overhead precision performance. As a worker gets physically fatigued, this overhead precision performance is expected to deteriorate. This indicates the need to evaluate the effect of a passive shoulder exoskeleton on development of physical fatigue, as well as the effect of such exoskeleton on overhead precision work when physically fatigued.
This PhD research aims to contribute to the rapidly evolving field of occupational exoskeletons. More specifically, the research line within this thesis focuses on exo skeleton assessment with four main research objectives: (1) providing a literature-based framework for occupational exoskeleton benchmarking, (2) investigating the effectiveness of two commercially available passive shoulder exoskeletons in real-life working situations, (3) evaluating the impact of the Exo4Work exoskeleton on simulated occupational overhead and non-overhead tasks, (4) investigating how passive shoulder exoskeleton support affects the reduction in task performance caused by physical fatigue.
To increase comparability among exoskeleton assessment studies, an assessment framework based on the insights gathered through an evidence mapping systematic re view was proposed in Chapter 2. This chapter recognizes the value of variability in assessment protocols in order to obtain an overall overview of the effect of exoskeletons on the users, but the presented framework strives to facilitate benchmarking the effect of occupational exoskeletons on the users across this variety of assessment protocols. The increased homogeneity and repeatability that arises from following this framework will expedite the establishment of the comprehensive research base on the effectiveness and usability of occupational exoskeletons. This may enable large-scale implementation of these exoskeletons in the industry.
In Chapter 3, the results showed that effects observed in a laboratory environment cannot be transferred to a field situation without caution. The muscle activity reducing effect of the commercially available exoskeletons in a controlled setting approximated effect reported in literature. In the field those effects were weaker, notwithstanding the workstations were specifically selected because of the high working height.
Chapter 3 emphasizes the need for more representative exoskeleton assessments. The recommendations in Chapter 2 contain the building blocks to conduct more representative assessments. Exoskeleton assessments focusing on the effect of the device on the user should recruit participants representing the population of industrial workers, including women and workers of age. In addition, researchers should first observe relevant workstations in the field, and mimic these scenarios in the laboratory. Insufficient robustness of a prototype may confine an optimally representative assessment experiment.
The outcomes from the preceding chapters were used to design the assessment protocols in Chapters 4 and 5. The first assessment of the novel passive shoulder exoskeleton, Exo4Work, investigated the effect of the exoskeleton on the user, and focused specifically on the design goals of this exoskeleton. The next experiment with this exoskeleton induced physical fatigue and explored how this physical fatigue interacted with exoskeleton support during precision overhead work.
The Exo4Work exoskeleton provided a limited amount of support, yet the exoskeleton reduced shoulder muscle activity and attenuated muscle fatigue development during overhead work in Chapter 4. Since the activity of the muscles working against the direction of support of the exoskeleton did not substantially increase, providing one third of the gravitational torque of the arms may represent a compromise between effective assistance while working above the head and resistance while lowering the arms. During non-overhead work, the hindrance of the exoskeleton was minimal, but participants perceived an increased level of frustration. Comfort at the level of the interfaces can also be improved. These results stress the potential of occupational shoulder exoskeletons in overhead working situations and may direct towards longitudinal field experiments. The last experiment of this PhD thesis (Chapter 5) was designed to evaluate how physical fatigue and exoskeleton support interacted during overhead precision work. In contrast to the hypothesis, support of the Exo4Work exoskeleton did not mitigate the fatigue-induced reduction in overhead precision performance. However, without exoskeleton support participants used compensatory movements to allow executing the task with smaller shoulder elevation angles. With support of the exoskeleton, the shoulde elevation angles were not affected by the physical fatigue. Those compensatory movements without exoskeleton were suggested to increase load on periarticular tissues, and may thus increase the risk of injury occurrence.
Overall, the research findings of this dissertation indicate that:
1. variability among exoskeleton assessment studies is large. The assessment frame work in Chapter 2 guides researchers to design an assessment protocol while in creasing comparability between studies.
2. most exoskeleton assessments were conducted in laboratories, but transferring those outcomes to the field is not obvious. Laboratory studies should include more representative participants and better mimic field situations.
3. the Exo4Work exoskeleton reduced shoulder muscle activity and development of muscle fatigue, while hindrance of the exoskeleton was limited. Providing one third of the gravitational torque of the arm may represent a compromise between effective assistance while working above the head and resistance while lowering the arms.
4. the use of a passive shoulder exoskeleton allows for overhead work without compensatory movements when physically fatigued. This suggests a passive shoulder exoskeleton may provide additional advantages in a physically fatigued state.
Future work could assess the effect of the Exo4Work exoskeleton in the field. Small modifications in the design will enable a better exoskeleton fit to a population of occupational workers. Another interesting path to explore is the subjective perception of the user while working with the exoskeleton. Insights into which parameters affect the subjective experience may contribute to the improvement of the prototype and could pave the path towards effective control of wearable robotics that is also appreciated by the wearer. Passive shoulder exoskeletons reduce acute physical load when working overhead, but these devices are only applicable for very specific scenarios. To expand the versatility of a shoulder exoskeleton, an active actuation system could be embedded in the device. Future research should confirm whether such active device effectively in creases versatility, and how this affects the user. Close collaboration between research in engineering and movement sciences can further exploit the potential of occupational exoskeletons, and create effective exoskeletons workers like to wear.