Hostname: page-component-78c5997874-ndw9j Total loading time: 0 Render date: 2024-11-05T02:41:23.978Z Has data issue: false hasContentIssue false

Occupational exoskeletons: A roadmap toward large-scale adoption. Methodology and challenges of bringing exoskeletons to workplaces

Published online by Cambridge University Press:  17 September 2021

Simona Crea*
Affiliation:
Scuola Superiore Sant’Anna, The BioRobotics Institute, Pontedera, Italy IRCCS Fondazione Don Gnocchi, Florence, Italy
Philipp Beckerle
Affiliation:
Chair of Autonomous Systems and Mechatronics, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany Institute for Mechatronic Systems, Technische Universität Darmstadt, Darmstadt, Germany
Michiel De Looze
Affiliation:
TNO, Leiden, The Netherlands
Kevin De Pauw
Affiliation:
Human Physiology and Sports Physiotherapy Research Group, and Brussels Human Robotics Research Center (BruBotics), Vrije Universiteit Brussel, Brussels, Belgium
Lorenzo Grazi
Affiliation:
Scuola Superiore Sant’Anna, The BioRobotics Institute, Pontedera, Italy
Tjaša Kermavnar
Affiliation:
School of Design, and Confirm Smart Manufacturing Centre, University of Limerick, Limerick, Ireland
Jawad Masood
Affiliation:
Processes and Factory of the Future Department, CTAG – Centro Tecnológico de Automoción de Galicia, Pontevedra, Spain
Leonard W. O’Sullivan
Affiliation:
School of Design, and Confirm Smart Manufacturing Centre, University of Limerick, Limerick, Ireland
Ilaria Pacifico
Affiliation:
Scuola Superiore Sant’Anna, The BioRobotics Institute, Pontedera, Italy
Carlos Rodriguez-Guerrero
Affiliation:
Robotics and Multibody Mechanics Research Group, Department of Mechanical Engineering, Vrije Universiteit Brussel and Flanders Make, Brussel, Belgium
Nicola Vitiello
Affiliation:
Scuola Superiore Sant’Anna, The BioRobotics Institute, Pontedera, Italy IRCCS Fondazione Don Gnocchi, Florence, Italy
Danijela Ristić-Durrant*
Affiliation:
Institute of Automation, University of Bremen, Bremen, Germany
Jan Veneman*
Affiliation:
Chair of COST Action 16116, Hocoma Medical GmbH, Zürich, Switzerland

Abstract

The large-scale adoption of occupational exoskeletons (OEs) will only happen if clear evidence of effectiveness of the devices is available. Performing product-specific field validation studies would allow the stakeholders and decision-makers (e.g., employers, ergonomists, health, and safety departments) to assess OEs’ effectiveness in their specific work contexts and with experienced workers, who could further provide useful insights on practical issues related to exoskeleton daily use. This paper reviews present-day scientific methods for assessing the effectiveness of OEs in laboratory and field studies, and presents the vision of the authors on a roadmap that could lead to large-scale adoption of this technology. The analysis of the state-of-the-art shows methodological differences between laboratory and field studies. While the former are more extensively reported in scientific papers, they exhibit limited generalizability of the findings to real-world scenarios. On the contrary, field studies are limited in sample sizes and frequently focused only on subjective metrics. We propose a roadmap to promote large-scale knowledge-based adoption of OEs. It details that the analysis of the costs and benefits of this technology should be communicated to all stakeholders to facilitate informed decision making, so that each stakeholder can develop their specific role regarding this innovation. Large-scale field studies can help identify and monitor the possible side-effects related to exoskeleton use in real work situations, as well as provide a comprehensive scientific knowledge base to support the revision of ergonomics risk-assessment methods, safety standards and regulations, and the definition of guidelines and practices for the selection and use of OEs.

Type
Review Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2021. Published by Cambridge University Press

Introduction

Occupational exoskeletons (OEs) can be defined as personal assistive devices that can reduce the physical load on workers performing demanding activities, by acting synergistically with their users (Monica et al., Reference Monica, Anastasi and Draicchio2020). Currently, most OEs are designed either to support specific body parts or to augment specific human capabilities. Regarding the former, the majority of OE devices have been developed to reduce the physical load on the upper limbs and the lumbar region of the spine, whereas support of other body parts, such as the knees or wrists, have been less explored.

As work-related musculoskeletal disorders (MSDs)Footnote 1 are associated with high costs to the employers, either related to direct compensation costs or indirect costs (e.g., lost wages, lost production, cost of recruiting and training replacement workers, and healthcare costs for rehabilitating the affected workers), companies have shown interest in OEs as possible solutions to reduce exposure of their workers to physical risk factors that can cause MSDs, and alternatives and/or complementary aids to more expensive solutions for reducing physical strain on their workers (such as collaborative robots). Hence, the main long-term expectations of companies on OEs are the reduced occurrence of work-related MSDs and related improved productivity (due to reduced absence rate) (Howard et al., Reference Howard, Murashov, Lowe and Lu2020; Monica et al., Reference Monica, Anastasi and Draicchio2020).

Motivated by the large potential to improve the ergonomics of workstations in their production lines, car manufacturers, in particular, have pioneered research in using the robots at workplaces. In this field, despite the extensive use of robots in most welding, painting, and assembly operations, final assembly lines still require several manual operations, such as the assembly tasks of the car interior, underbody, doors, windows or trims, and sealing of the doors. These are performed at a high pace (work cycles of a few minutes for small cars), and in some cases require maintaining awkward postures while performing high-precision actions. Recently, several car manufacturers have evaluated and tested exoskeletons within their plants, in some cases developing custom devices to meet their specific needs (Spada et al., Reference Spada, Ghibaudo, Gilotta, Gastaldi and Cavatorta2017; Gillette and Stephenson, Reference Gillette and Stephenson2018; Claramunt-Molet et al., Reference Claramunt-Molet, Domingo-Mateu, Danús-Jaume, Ugartemendia-Etxarri, De La Maza, Enriquez-Carrera, Muñoz-Fernandez, Domingo, Miralles, Font-Llagunes and Idelsohn-Zielonka2019; Ferreira et al., Reference Ferreira, Gaspar, Fujão, Nunes and Nunes2020; Iranzo et al., Reference Iranzo, Piedrabuena, Iordanov, Martinez-Iranzo and Belda-Lois2020). In addition to the automotive field, exoskeletons have been proposed in various other manufacturing contexts, as well as in nonmanufacturing domains, such as logistics, construction, agriculture, and patient care. However, currently, the actual adoption of exoskeletons in real scenarios is limited compared to the expectations.

Such a limited-scale adoption of OEs may be related to different factors. One of them is the lack of clear evidence of adequate effectiveness of the devices in the final workplaces. While system capability to reduce the load on specific body parts in stereotyped gestures could be investigated by OE manufacturers in laboratory settings (in-lab), the device effectiveness is use-case specific and therefore should be studied in dedicated experiments: use-case specific experiments could be carried out in laboratory-reconstructed workstations or directly at the individual workstation (in-field). Performing product-specific validation studies in-field would allow the stakeholders and decision-makers to assess the effectiveness of devices in their specific work contexts and with experienced workers who could provide useful insights into practical issues related to exoskeleton use in daily practice. In such field studies, effectiveness indices may include experience-related indicators, measured through questionnaires and structured interviews, as well as instrumental parameters, based on kinematics, kinetics, and physiological (e.g., cardiovascular, pulmonary, and muscular) measures (Ármannsdóttir et al., Reference Ármannsdóttir, Beckerle, Moreno, van Asseldonk, Manrique-Sancho, Del-Ama, Veneman and Briem2020; Torricelli et al., Reference Torricelli, Rodriguez-Guerrero, Veneman, Crea, Briem, Lenggenhager and Beckerle2020).

This paper aims to provide a comprehensive overview of the scientific methods used to investigate the effectiveness of OEs, comparing the methods used in lab and field conditions. Also, the paper presents the position of the authors suggesting a roadmap that might support large-scale adoption of the technology. The following sections are organized as follows. Section “Technological Approaches to Design OEs” introduces different OE technologies, with a particular focus on their classification based on different kinematics and actuation approaches, and the regulatory-based classification, since these aspects are relevant, together with other aspects, for assessing the effectiveness of the devices in different contexts. Section “Scientific Evidence of OE Effectiveness in Laboratory and Field Studies” provides a narrative overview of the state-of-the-art of methodological approaches to validate exoskeletons in laboratory and field studies. Finally, section “Roadmap Toward Large-Scale Adoption of OEs” outlines a proposed roadmap for large-scale adoption of OEs, including the process of implementing OEs into specific workplaces, communication with different stakeholders, and future perspectives.

Technological Approaches to Design OEs

Industrial scenarios and workflows tend to be very dynamic and diverse, and workers rarely continuously repeat a limited set of movements. Taking this into account, OEs should be designed to comply with very diverse conditions of use while finding the most appropriate trade-offs to fulfill various requirements (e.g., lightweight design and system adaptability). This is a challenge for exoskeletons designers. Several design architectures have been proposed to realize compact, lightweight, and effective OEs by following diverse approaches, mostly in terms of kinematic structure and type of actuation (de Looze et al., Reference de Looze, Bosch, Krause, Stadler and O’Sullivan2016; Fox et al., Reference Fox, Aranko, Heilala and Vahala2019; Toxiri et al., Reference Toxiri, Näf, Lazzaroni, Fernández, Sposito, Poliero, Monica, Anastasi, Caldwell and Ortiz2019; Gull et al., Reference Gull, Bai and Bak2020; Huysamen et al., Reference Huysamen, Power and O’Sullivan2020; Kermavnar et al., Reference Kermavnar, de Vries, de Looze and O’Sullivan2020). The classification of OEs in terms of kinematics structure and actuation is illustrated in Figure 1.

Figure 1. Occupational exoskeletons classification based on kinematic structure (a) and type of actuation (b). Concerning the kinematics, rigid-structure devices can be classified in anthropomorphic and nonanthropomorphic devices, whereas soft exosuits do not present any kinematic structure. Actuation types include passive, semi-active, and active systems. In the picture actuation types are depicted for an anthropomorphic device. Sketches are provided for upper-limb devices.

The kinematic compatibility of a wearable device is paramount for a comfortable fit of a wearable device, effective torque transfer, and minimization of undesired forces to the human skeletal system. The exoskeleton should allow the user to move within his/her physiological range of movement and must account for the translations associated with the joint rotational movements (Voilqué et al., Reference Voilqué, Masood, Fauroux, Sabourin and Guezet2019; Rossini et al., Reference Rossini, Geeroms, Lefeber and Rodriguez-Guerrero2021). Rigid kinematic structures of upper-limb OEs typically originate from a frame attached to the pelvic region of the human body and end at the arm attachment point. Lower-back exoskeletons, instead, usually have rigid kinematic structures surrounding the user’s hips and lower back, with attachment points located in correspondence with the thighs and the trunk. Based on their kinematic structures, rigid exoskeletons can be classified as anthropomorphic or nonanthropomorphic (e.g., end-effector-like) devices (Schiele and van der Helm, Reference Schiele and van der Helm2006; Naf et al., Reference Naf, Junius, Rossini, Rodriguez-Guerrero, Vanderborght and Lefeber2018). Exoskeletons with anthropomorphic kinematic structures include robotic joints that need to be aligned with the user’s joint axes, thus misalignment-compensation strategies should be included to counteract the effects of axis misalignments; whereas exoskeletons with nonanthropomorphic kinematic structures do not require a direct correspondence between the robot’s and user’s axes of rotation (Nef et al., Reference Nef, Guidali and Riener2009; Accoto et al., Reference Accoto, Sergi, Tagliamonte, Carpino, Sudano and Guglielmelli2014). Regarding upper-limb exoskeletons, anthropomorphic kinematic structures typically have the shoulder-elevation axis collocated in correspondence with the shoulder glenohumeral joint, and related misalignment compensation strategies based on passive degrees of freedom and/or compliant elements. Back-support exoskeletons typically exert assistive force around the human hips and the lumbosacral joint (i.e., L5/S1 joint), respectively with anthropomorphic and nonanthropomorphic approaches due to the difficulty in identifying and accessing the L5/S1 joint specifically. Different from rigid-structure devices, soft exosuits are wearable clothing-like devices that can generate moments around biological joints through pulling cables and textiles acting in parallel to the action of muscles and tendons. In these systems compressing loads are not sustained by any external rigid structure but are sustained by the wearer’s bone structure. Lumbar exosuits apply assistive forces in parallel with the lumbar and hip extensor muscles, with elastic textiles and/or cables running along the back and thighs (Lamers and Zelik, Reference Lamers and Zelik2021), whereas upper-limb exosuits assist the shoulder and/or elbow flexors (Kim et al., Reference Kim, Little, Noronha, Xiloyannis, Masia and Accoto2020a).

As far as actuation is concerned, OEs can be classified into three subcategories: active, passive, or semi-active exoskeletons. Active exoskeletons use powered actuators to generate assistive torque and rely on sensors and control units to synchronize robot action with the user’s motion, whereas passive exoskeletons exploit springs or spring-like elements to store and release energy in various phases of the human movement (de Looze et al., Reference de Looze, Bosch, Krause, Stadler and O’Sullivan2016). The comparison of these two actuation alternatives should account for multiple factors. Currently, passive OEs have higher maturity compared to their active counterparts, as their functioning does not involve the use of actuators, batteries, wiring, and electronics and their physical human–robot interface has a repeatable and intuitive behavior. In very dynamic and diverse operating environments, active OEs can be more adaptable than passive devices in different tasks, but the need for extremely accurate control algorithms currently prevents their use in field scenarios (De Bock et al., Reference De Bock, Ghillebert, Govaerts, Elprama, Marusic, Serrien, Jacobs, Geeroms, Meeusen and De Pauw2020). To increase system adaptability and with the final objective to further increase the system acceptance, mechanical clutches and null-assistance configurations have been implemented into some passive devices: they usually require some type of human intervention to engage or disengage the mechanisms to allow for larger freedom of movement while not executing the specific tasks they were designed for (e.g., walking or grasping tools) (Baltrusch et al., Reference Baltrusch, van Dieen, van Bennekom and Houdijk2020c). This may end up being cumbersome and impractical in the field, especially when manual tasks are involved. In the attempt to achieve the trade-off between the larger adaptability of active systems and greater usability of passive devices, a trade-off could be represented by semi-active exoskeletons that use low-power servo motors to adapt the behavior of the device based on the user’s needs, for example, by adapting the level of assistance or engaging/disengaging the actuation mechanisms (Figure 1).

In addition to considerations of the actuation architecture, the comparison of the actuation of different OEs should consider the amount of assistive torque the devices can exert on the target body parts. The ratio between the assistive torque and the physiological joint torque exerted by muscles in a specific task could be computed to provide an estimate of the biomechanical unloading due to the use of the device. The physiological torque exerted by lower-back extensor muscles around the L5-S1 joint can reach up to 200 Nm during trunk flexion and extension movements without handling tools, with an angular velocity in the range of about 90–120 deg/s (and peak power of 0.7–1.1 W/kg) (Saraceni et al., Reference Saraceni, Campbell, Kent, Ng, Straker and O’Sullivan2021), or up to 300 Nm when lifting a 15-kg load (Koopman et al., Reference Koopman, Kingma, Faber, de Looze and van Dieën2019a), whereas shoulder elevator muscles typically exert 30 Nm when keeping the arms overhead holding tools such as a drill (Anton et al., Reference Anton, Shibley, Fethke, Hess, Cook and Rosecrance2001). To provide sufficient assistive torque to reduce the physiological workload on the lumbar or shoulder joints, most devices are designed to output torques in the range of 10–20 Nm per hip or up to 40 Nm at the back (Huysamen et al., Reference Huysamen, de Looze, Bosch, Ortiz, Toxiri and O’Sullivan2018b; von Glinski et al., Reference von Glinski, Yilmaz, Mrotzek, Marek, Jettkant, Brinkemper, Fisahn, Schildhauer and Geßmann2019; Lanotte et al., Reference Lanotte, McKinney, Grazi, Chen, Crea and Vitiello2021) and 3–6 Nm for the shoulder (de Vries et al., Reference de Vries, Murphy, Könemann, Kingma and de Looze2019; Grazi et al., Reference Grazi, Trigili, Proface, Giovacchini, Crea and Vitiello2020), spanning between the 5 and 20% of the physiological joint torque.

Example of state-of-the-art OEs classified according to their kinematics or actuation is given in Table 1.

Table 1. Classification of occupational exoskeletons (OEs) according to kinematics, structure, and actuation; within each category, a nonexhaustive list of commercial and prototypical OEs is given

Scientific Evidence of OE Effectiveness in Laboratory and Field Studies

A search of electronic and personal databases was performed to identify papers describing objective and subjective evaluations of OEs in laboratory and field studies. Methodological aspects of exoskeleton testing were reviewed for active and passive back-support and upper-limb devices. In total, 99 studies were analyzed and are listed in the Supplementary Materials. A summary of studied tasks and effects of exoskeleton use is provided below.

Laboratory Studies

The main goal of laboratory studies is to validate OEs, and to investigate their biomechanical effects on the users’ physical effort, or the level of assistance they can provide. An additional goal of laboratory studies is to identify potentially undesired effects (e.g., due to load redistribution on the human body) during the tasks under investigation. Biomechanical studies carried out in laboratory conditions can benefit from the use of motion tracking systems, force platforms, or high-density EMG recorders, which would be impractical to use outside the laboratory.

The potential of back-support and upper-limb exoskeletons to reduce the mechanical loads in the back and shoulder regions has been frequently addressed in laboratory studies (McFarland and Fischer, Reference McFarland and Fischer2019; Theurel and Desbrosses, Reference Theurel and Desbrosses2019; de Vries and de Looze, Reference de Vries and de Looze2020; Howard et al., Reference Howard, Murashov, Lowe and Lu2020; Kermavnar et al., Reference Kermavnar, de Vries, de Looze and O’Sullivan2020).

Back-support exoskeletons

A total of 57 laboratory studies of back-support exoskeletons were identified (Supplementary Table S5).

The devices were typically tested during static bending or lifting tasks (including single lifts and repetitive lifting), and less often during complex tasks that more closely resembled real-work situations.

Studies of static bending with OEs tended to involve passive devices, tested on healthy young or middle-aged participants who were predominantly male. The bending position was typically determined by trunk flexion angle (40–55°; Bosch et al., Reference Bosch, van Eck, Knitel and de Looze2016; Wei et al., Reference Wei, Shijia, Yuxuan, Jihua and Xichuan2020a), or hand distance from the floor (Kim et al., Reference Kim, Madinei, Alemi, Srinivasan and Nussbaum2020b). The bending duration was either decided in advance, ranging from 5 to 300 s (Koopman et al., Reference Koopman, Näf, Baltrusch, Kingma, Rodriguez-Guerrero, Babič, Kingma, Rodriguez-Guerrero, Babič, de Looze and van Dieën2020a; Wei et al., Reference Wei, Shijia, Yuxuan, Jihua and Xichuan2020a, Reference Wei, Wang, Qu, Gu, Lin and Yue2020b), or determined by the participants’ perception of discomfort (Bosch et al., Reference Bosch, van Eck, Knitel and de Looze2016). The studied effects most commonly included EMG activity of back, abdominal, upper- and lower-limb muscles, metabolic cost, various joint angles, and ground-reaction forces (Bosch et al., Reference Bosch, van Eck, Knitel and de Looze2016; Koopman et al., Reference Koopman, Kingma, Faber, de Looze and van Dieën2019a, Reference Koopman, Näf, Baltrusch, Kingma, Rodriguez-Guerrero, Babič, Kingma, Rodriguez-Guerrero, Babič, de Looze and van Dieën2020a; Kim et al., Reference Kim, Madinei, Alemi, Srinivasan and Nussbaum2020b; Wei et al., Reference Wei, Shijia, Yuxuan, Jihua and Xichuan2020a, Reference Wei, Wang, Qu, Gu, Lin and Yue2020b). Among subjective measures, local discomfort, perception of balance, device fit, comfort, body movement constraints, and overall helpfulness of the OE were assessed (Bosch et al., Reference Bosch, van Eck, Knitel and de Looze2016; Kim et al., Reference Kim, Madinei, Alemi, Srinivasan and Nussbaum2020b).

Single symmetric and asymmetric lifts were performed with active and passive back-support exoskeletons, and also predominantly involved healthy young and middle-aged male participants. OE target users in the luggage-handling sector of the airline industry were included in one study (Koopman et al., Reference Koopman, Näf, Baltrusch, Kingma, Rodriguez-Guerrero, Babič, Kingma, Rodriguez-Guerrero, Babič, de Looze and van Dieën2020a). The starting position of the loads ranged from the floor to 10 cm above the knee, and they were typically lifted to waist height, in many cases including trunk rotation (Chen et al., Reference Chen, Grazi, Lanotte, Vitiello and Crea2018; Huysamen et al., Reference Huysamen, de Looze, Bosch, Ortiz, Toxiri and O’Sullivan2018b; Lanotte et al., Reference Lanotte, Grazi, Chen, Vitiello and Crea2018; Toxiri et al., Reference Toxiri, Koopman, Lazzaroni, Ortiz, Power, de Looze, O’Sullivan and Caldwell2018; Alemi et al., Reference Alemi, Geissinger, Simon, Chang and Asbeck2019; Goršič et al., Reference Goršič, Regmi, Johnson, Dai and Novak2019; Koopman et al., Reference Koopman, Toxiri, Power, Kingma, van Dieën, Ortiz and de Looze2019b, Reference Koopman, Näf, Baltrusch, Kingma, Rodriguez-Guerrero, Babič, Kingma, Rodriguez-Guerrero, Babič, de Looze and van Dieën2020a, Reference Koopman, Kingma, de Looze and van Dieën2020b; Picchiotti et al., Reference Picchiotti, Weston, Knapik, Dufour and Marras2019; Hussain et al., Reference Hussain, Park, Kim, Kim and Lee2020; Hyun et al., Reference Hyun, Lim, Park and Nam2020). Loads ranged from 5 through 18 kg across the studies and were lifted using different techniques at a comfortable pace, or every 6–15 s. The EMG activity of back, abdominal and lower-limb muscles were most investigated, together with various joint angles, ground-reaction forces, lower-back extension moments and/or compression, trunk extension time, and contact pressure. Among subjective measures, perceived exertion, local discomfort, musculoskeletal pressure, and exoskeleton usability were assessed in some studies (Huysamen et al., Reference Huysamen, de Looze, Bosch, Ortiz, Toxiri and O’Sullivan2018b; Alemi et al., Reference Alemi, Geissinger, Simon, Chang and Asbeck2019).

Repetitive lifting was performed with active and passive exoskeletons, mainly by healthy young and middle-aged participants, predominantly men (Baltrusch et al., Reference Baltrusch, van Dieën, Bruijn, Koopman, van Bennekom and Houdijk2019; Heo et al., Reference Heo, Kim and Kim2020; Madinei et al., Reference Madinei, Alemi, Kim, Srinivasan and Nussbaum2020a, Reference Madinei, Alemi, Kim, Srinivasan and Nussbaum2020b; Wei et al., Reference Wei, Shijia, Yuxuan, Jihua and Xichuan2020a, Reference Wei, Wang, Qu, Gu, Lin and Yue2020b). One study included workers with low-back pain (Baltrusch et al., Reference Baltrusch, Houdijk, van Dieën and Kruif2020a). Across the studies, loads of 6–20 kg were lifted during standing or kneeling, with the starting position ranging from the floor to knee level, and end position from 50 cm above the floor to hip height. Lifting was performed using different lifting techniques at 5–30 cycles/min, and lasted for 1–5 min or until exhaustion. Measured were the EMG activity of back, abdominal, upper- and lower-limb muscles, various joint angles, ground-reaction forces, number of lifting cycles until exhaustion, and metabolic cost (Baltrusch et al., Reference Baltrusch, van Dieën, Bruijn, Koopman, van Bennekom and Houdijk2019; Tan et al., Reference Tan, Kadone, Miura, Abe, Koda, Yamazaki, Sankai and Suzuki2019; Baltrusch et al., Reference Baltrusch, van Dieën, Koopman, Näf, Rodriguez-Guerrero, Babič and Houdijk2020b; Heo et al., Reference Heo, Kim and Kim2020; Madinei et al., Reference Madinei, Alemi, Kim, Srinivasan and Nussbaum2020a, Reference Madinei, Alemi, Kim, Srinivasan and Nussbaum2020b; Wei et al., Reference Wei, Shijia, Yuxuan, Jihua and Xichuan2020a, Reference Wei, Wang, Qu, Gu, Lin and Yue2020b). Subjective measures included perceived exertion, fatigue, balance, local discomfort, device fit, comfort, movement constraints, and overall usability (Tan et al., Reference Tan, Kadone, Miura, Abe, Koda, Yamazaki, Sankai and Suzuki2019; Madinei et al., Reference Madinei, Alemi, Kim, Srinivasan and Nussbaum2020a, Reference Madinei, Alemi, Kim, Srinivasan and Nussbaum2020b).

The literature demonstrates a very wide range of approaches to experimental design for assessing back-support exoskeletons. This variability among studies makes it difficult to directly compare the results. The first attempt toward a standardized testing approach was made with the development of a functional performance test battery which includes 12 tasks to assess the support and hindrance of the user by the exoskeleton (Baltrusch et al., Reference Baltrusch, van Dieën, van Bennekom and Houdijk2018). The tasks supported by the exoskeleton include repetitive lifting (20-kg load from ankle height, 2 min), carrying (20-kg box, 10 m), static bending (a simple manual task at knee height, 30–60° trunk flexion, max. 5 min), and three-point kneeling (a simple manual task with one hand on the floor, max. 5 min). The tasks hindered by the exoskeleton include walking, sit-to-stand/stand-to-sit, climbing stairs and ladder. The tasks where the exoskeleton potentially affects the range of motion involve forward bending with extended knees, wide stance, trunk rotation, and squats.

Other testing scenarios have also been reported involving healthy, predominantly young participants. Passive back-support exoskeletons were tested during walking on a treadmill for 5 min or across the laboratory at preferred speed (Baltrusch et al., Reference Baltrusch, van Dieën, Bruijn, Koopman, van Bennekom and Houdijk2019; Goršič et al., Reference Goršič, Regmi, Johnson, Dai and Novak2019), and the EMG activity of back, abdominal, and lower-limb muscles was measured, as well as low-back extension moment and various joint angles (Baltrusch et al., Reference Baltrusch, van Dieën, Bruijn, Koopman, van Bennekom and Houdijk2019; Goršič et al., Reference Goršič, Regmi, Johnson, Dai and Novak2019). During standing up from a stool, the EMG activity of back and abdominal muscles, low-back extension moment, and trunk flexion angle were measured (Goršič et al., Reference Goršič, Regmi, Johnson, Dai and Novak2019). In sitting with perturbation from the front, back, left and right, trunk deflection and EMG activity of back and abdominal muscles were measured (Goršič et al., Reference Goršič, Regmi, Johnson, Dai and Novak2019). During repetitive pick-and-place tasks at 40° trunk flexion with occasional lifting of bins to shoulder height, the EMG activity of back, abdominal, and upper-limb muscles was measured, and local discomfort was assessed (Bosch et al., Reference Bosch, van Eck, Knitel and de Looze2016). In some cases, the laboratory settings were a mock-up of the workplace environment (Baltrusch et al., Reference Baltrusch, van Dieën, Koopman, Näf, Rodriguez-Guerrero, Babič and Houdijk2020b), or simulated assembly work (Bosch et al., Reference Bosch, van Eck, Knitel and de Looze2016).

Active exoskeletons were studied during snow shoveling as fast as possible until fatigued (Miura et al., Reference Miura, Kadone, Koda, Abe, Endo, Murakami, Doita, Kumagai, Nagashima, Fujii, Noguchi, Funayama, Kawamoto, Sankai and Yamazaki2018), and during a task composed of sit-to-stand, walking, lifting–lowering, walking back, and stand-to-sit (Chen et al., Reference Chen, Grazi, Lanotte, Vitiello and Crea2018). In the former case, the number of scoops, shoveling time and distance, heart rate, blood pressure, and lumbar fatigue perception were measured; and in the latter case, the EMG activity of back muscles was assessed.

Upper-limb exoskeletons

A total of 29 laboratory studies of upper-limb exoskeletons (Supplementary Table S3) were identified and analyzed.

Across the studies performed in a laboratory environment, the majority of exoskeletons subjected to research were passive while a few studies involved active exoskeletons.

Tasks performed in laboratories to investigate the effects of wearing an exoskeleton vary across studies. Most studies concentrate on tasks requiring arm elevation with varying elevation angles (roughly ranging from hands at waist level, in front of the body, up to overhead work) (de Vries et al., Reference de Vries, Murphy, Könemann, Kingma and de Looze2019). The main focus across the studies is on static tasks where the arms are elevated and held in a static position (Huysamen et al., Reference Huysamen, Bosch, de Looze, Stadler, Graf and O’Sullivan2018a, Schmalz et al., Reference Schmalz, Schändlinger, Schuler, Bornmann, Schirrmeister, Kannenberg and Ernst2019; Grazi et al., Reference Grazi, Trigili, Proface, Giovacchini, Crea and Vitiello2020), or on quasi-static tasks, where a specific activity is performed with hardly any arm movement around the shoulder joint (Pacifico et al., Reference Pacifico, Scano, Guanziroli, Moise, Morelli, Chiavenna, Romo, Spada, Colombina, Molteni, Giovacchini, Vitiello and Crea2020). These quasi-static tasks comprise (overhead) assembly (Perez Luque et al., Reference Perez Luque, Högberg, Iriondo Pascual, Lämkull and Garcia Rivera2020), drilling (Alabdulkarim and Nussbaum, Reference Alabdulkarim and Nussbaum2019), wiring (Kim et al., Reference Kim, Nussbaum, Mokhlespour Esfahani, Alemi, Alabdulkarim and Rashedi2018), and sanding (Moyon et al., Reference Moyon, Poirson and Petiot2018). The handled loads (e.g., hand tools) in these types of tasks range from 0 to 5 kg. More dynamic tasks that have been studied are carrying and stacking (Theurel et al., Reference Theurel, Desbrosses, Roux and Savescu2018) and plastering of ceilings and walls (de Vries et al., Reference de Vries, Krause and de Looze2021).

Various variables have been used to assess the direct effect on shoulder load and the short-term human body response. For assessing the effect on workload, a variable that has been addressed in lab studies is the shoulder torque (e.g., de Vries et al., Reference de Vries, Murphy, Könemann, Kingma and de Looze2019), that is, the torque to be generated by the person when wearing an exoskeleton, estimated by inverse dynamics techniques. Levels of muscle activation have also been used in multiple studies as a parameter to evaluate the effect of an exoskeleton on shoulder load (e.g., Naito et al., Reference Naito, Obinata, Nakayama and Hase2007; Huysamen et al., Reference Huysamen, Bosch, de Looze, Stadler, Graf and O’Sullivan2018a; Kim et al., Reference Kim, Nussbaum, Mokhlespour Esfahani, Alemi, Alabdulkarim and Rashedi2018; Van Engelhoven et al., Reference Van Engelhoven, Poon, Kazerooni, Barr, Rempel and Harris-Adamson2018). Muscles activation has been assessed through EMG measurements. EMG measures were carried out on the arm elevator muscles, particularly the anterior deltoid, medial deltoid, trapezius pars descendens, and biceps brachii muscles, extensor muscles, like the posterior deltoid and triceps brachii, and shoulder stabilizers, such as the pectoralis major muscle. In addition to average activation, the assessment of muscle fatigue by frequency analysis has been considered in a few laboratory studies, despite the difficulty of controlling all possible confounding factors (Rashedi et al., Reference Rashedi, Kim, Nussbaum and Agnew2014). Other variables have also been studied: endurance was studied by Spada et al. (Reference Spada, Ghibaudo, Gilotta, Gastaldi and Cavatorta2017), various kinematic variables (Maurice et al., Reference Maurice, Čamernik, Gorjan, Schirrmeister, Bornmann, Tagliapietra, Latella, Pucci, Fritzsche, Ivaldi and Babič2019; Schmalz et al., Reference Schmalz, Schändlinger, Schuler, Bornmann, Schirrmeister, Kannenberg and Ernst2019; Grazi et al., Reference Grazi, Trigili, Proface, Giovacchini, Crea and Vitiello2020; Pacifico et al., Reference Pacifico, Scano, Guanziroli, Moise, Morelli, Chiavenna, Romo, Spada, Colombina, Molteni, Giovacchini, Vitiello and Crea2020) or physiological variables like heartrate or oxygen uptake (Theurel et al., Reference Theurel, Desbrosses, Roux and Savescu2018; Maurice et al., Reference Maurice, Čamernik, Gorjan, Schirrmeister, Bornmann, Tagliapietra, Latella, Pucci, Fritzsche, Ivaldi and Babič2019; Grazi et al., Reference Grazi, Trigili, Proface, Giovacchini, Crea and Vitiello2020) have been investigated by researchers in different studies.

Field Studies

The goal of exoskeleton evaluation in field studies is to determine their effects in the workplace settings on one or more of the following variables: (a) behavior, (b) use, usability, and acceptance, (c) performance, and (d) workload, discomfort, and fatigue. As the use of complex experimental equipment at the workplace would be impractical, instrumental measurement systems typically include wearable sensors, such as IMUs and wireless EMG recorders.

While studying the incidence of work-related MSDs (e.g., low back pain and neck-shoulder injury) could be considered a primary goal of field studies, it is important to understand that large long-term controlled studies have not been performed yet, likely due to the significant effort required and relatively low maturity of the market.

Back-support exoskeletons

In total, five field studies of back-support exoskeletons were identified (Supplementary Table S6).

Graham et al. (Reference Graham, Agnew and Stevenson2009) performed a study of the Personal Lift Assist Device (PLAD) on healthy automotive operators during a highly standardized automotive assembly task. Two hours of 55-s cycles of manual work were monitored, with static bending and trunk rotation, standing, walking, and carrying of car parts. EMG activity of back and abdominal muscles, and joint angles were recorded.

The study by Amandels et al. (Reference Amandels, het Eyndt, Daenen and Hermans2019) of Laevo V2.4 was performed on middle-aged male shop-floor workers at a press and shear department with low-back pain. The participants were monitored 30 min per day for 3 weeks while performing their typical work tasks of frequent far-reaching with bending. Measures included EMG activity of the back, upper and lower-limb muscles, joint angles, perceived level of local discomfort, and users’ impression.

A 4-week field study of Laevo was performed by Hensel and Keil (Reference Hensel and Keil2019) on 30 male employees in the automotive industry. The study was conducted at five workplaces in the assembly and press shop (ground screw connection footwell, trunk insulation, installation cable harness, maintenance, and press set up) with tasks performed in a marked static forward bend. In addition, three workplaces with considerable upper-body flexion in logistics were selected to evaluate dynamic-repositioning activities. Perceived relief at the lower back and discomfort, usability, and user acceptance were evaluated.

Motmans et al. (Reference Motmans, Debaets and Chrispeels2019) performed a study of Laevo V2.5 on middle-aged male order-pickers of cheese with low-back pain. The participants’ EMG activity of back and upper-limb muscles, joint angles, and perceived device effect on work task, effectiveness, and comfort were measured during 1.5 hr of cheese picking.

Settembre et al. (Reference Settembre, Maurice, Paysant, Theurel, Claudon, Kimmoun, Levy, Hani, Chenuel and Ivaldi2020) performed a field study of Laevo v1 on two male workers at a COVID-19 intensive care unit. The participants were equipped with the exoskeleton in compliance with the strict hygiene measures of the intensive care unit during the outbreak. During a typical 3-hr shift, each participant performed 10 prone positioning/supine positioning maneuvers that involved static forward bending to secure the patient’s head, forward trunk flexion during SP and forward trunk flexion during repositioning of the patient on the bed. The participants’ perceived physical effort and relief were measured, as well as their cardiac activity.

Upper-limb exoskeletons

In total, eight field studies of upper-limb exoskeletons were identified (Supplementary Table S4).

Gillette and Stephenson (Reference Gillette and Stephenson2018) conducted a field study of Levitate Airframe during an overhead automotive assembly at Toyota Canada. Eleven experienced male workers performed 10 automotive assembly tasks. Measurements included EMG activity, fatigue, and injury hazard.

Moyon et al. (Reference Moyon, Poirson and Petiot2018) conducted a field study of the upper-body passive exoskeleton Skelex at a boat-manufacturer worksite. Nine workers performed one of the finishing operations on a catamaran hull: sanding, coating, polishing, or painting. The measurements of heart rate, posture analysis, and usability were performed.

Gillette and Stephenson (Reference Gillette and Stephenson2019) conducted a field study of Levitate Airframe on six workers performing tasks that involved prolonged elevated arm postures, including cab assembly, hydraulic assembly, parts painting, parts hanging, and frame welding tasks. The EMG activity of arm, shoulder, and lower-back muscles, and fatigue hazard were measured.

Smets (Reference Smets2019) conducted a field trial of EksoVest with 22 operators during overhead automotive assembly work. The study was performed in three phases: the first consisted of the collection of subjective first impressions and feedback on proto-EksoVest (eight operators); the second consisted of estimating the usability (10 operators); and the third included a 3-month use for the collection of usability and musculoskeletal discomfort evaluations (four operators).

De Bock et al. (Reference De Bock, Ghillebert, Govaerts, Elprama, Marusic, Serrien, Jacobs, Geeroms, Meeusen and De Pauw2020) performed a field study in which four operators tested two commercial upper-limb exoskeletons (i.e., the ShoulderX and Skelex). The tasks consisted of transferring wind-screens from a trailer into a storage rack and then back to the trailer and transferring the wind-screens from the forklift platform to the storage rack and then back to the forklift platform. Muscular activity, heart rate, local discomfort, usability, and task load were assessed.

Hefferle et al. (Reference Hefferle, Snell and Kluth2020) performed a field study of two different overhead exoskeletons, Crimson Dynamics and Skelex V1, with eight male associates on the assembly line of an automotive manufacturer. The workers performed their daily tasks at the workstation with a mean shoulder joint angle of 90° depending on employee stature. Perceived local strain of shoulders and neck, and whole-body strain were assessed.

Iranzo et al. (Reference Iranzo, Piedrabuena, Iordanov, Martinez-Iranzo and Belda-Lois2020) performed a field study of Levitate Airframe on 12 workers (11 male, 1 female). The participants performed repetitive overhead work, assembling the car body from underneath the car with pneumatic screwdrivers. EMG activity was measured, as well as the angles of the neck, back, and upper-limb joints.

Wang et al. (Reference Wang, Le and Lin2021) performed a field study of the passive upper-limb exoskeleton PULE with 8 farmers at the orchard farming test site. Four participants performed fruit thinning with the arms in a static elevated position (100° shoulder flexion), and four participants sprayed pesticide with the hands swinging from 0° to 100° and back. EMG activity and the angle of the upper limb were measured.

Main methodological differences between laboratory and field studies

Figure 2 visualizes the metrics of back-support and upper-limb exoskeletons in laboratory and field conditions. Relatively few field studies have been conducted compared to laboratory studies, with back-support studies showing a ratio between field to laboratory studies equal to 1 to more than 10, and upper-limb studies showing a ratio of 1 to more than 3.

Figure 2. Frequency analysis of the objective and subjective metrics between laboratory and field studies for back-support and upper-limb exoskeletons. Acronyms: Technology Acceptance Model (TAM), NASA task load index (NASA TLX).

EMG and subjective measures (Rating of Perceived Exertion—RPE; System Usability Scale—SUS) are the most consistent metrics across field and laboratory studies. Indeed, the key measurement in laboratory studies is surface EMG signals are of interest for two reasons. Firstly, the forces that are generated by muscles around a specific joint directly lead (through action–reaction) to an increase in internal loading forces (compression and shear forces) on joint structures. Secondly, the level of muscle activation that is required for a specific task is a predictor of muscle fatigue development. Fatigue, in turn, could be assessed by a simultaneous temporal increase of the EMG amplitude and decrease of its frequency content (Basmajian and De Luca, Reference Basmajian and De Luca1985).

Another important parameter in the assessment of OEs is endurance, that is, the duration of a person’s capability to perform a specific task without exhaustion or before the onset of a predefined level of fatigue, evaluated by the perceived exertion on an RPE scale (Borg, Reference Borg1982). Also, perceived effort, discomfort, and fatigue have been frequently used to evaluate OEs. Perceived discomfort and fatigue (in contrast with perceived effort) are variables that need time to develop and therefore require relatively long trials to get relevant results. The selection of metrics depends on whether the studies are conducted in laboratory or field conditions, with a tendency toward more instrumental data gathering in laboratory conditions.

Although studies in controlled laboratory environments involve multidisciplinary approaches to provide the bigger picture of exoskeleton functioning, field studies mainly include easy-to-monitor and easy-to-use measures. Field studies are of major importance to prove whether or not results from laboratory studies are generalizable to the field conditions. From a practical point of view, this is not straightforward, since body movements in occupational settings are task-specific and thus differ from simulated stereotyped tasks, such as static bending for a given amount of time, lifting a box using a specific lifting technique, or walking 10 m on level ground without obstacles. Additionally, in the field studies, the effectiveness of exoskeletons is also challenged beyond the sole functionality of the device, which compromises the transfer of research results from the laboratory to the field conditions. Current wearable sensors for biomechanics measurements may still be too complex to be used in field conditions. As it would increase the complexity to the setup of the study and the experimental procedures (e.g., due to the need for frequent calibration procedures), future advancements in wearable sensory systems (such as inertial measurement devices, portable pressure-sensitive insoles, mobile electro-physiological devices, and gas exchange devices) could enable gathering more instrumental data also in field studies.

In addition to gathering data, the assessment of the measurement systems and metrics extracted from lab and field studies revealed that only in a few studies, instrumental and perception-related assessment of OEs efficacy served as the basis for the evaluation on how the risk factors associated with a certain task changed. In particular, for back-support OEs, only 13 out of 62 lab and field studies used biomechanics measurements to assess changes in the risk factors that have been commonly associated with low-back injury, that is, the cumulative low-back load, the peak low-back load and the flexion relaxation, or a subset of them (McGill and Kippers, Reference McGill and Kippers1994; Norman et al., Reference Norman, Wells, Neumann, Frank, Shannon and Kerr1998). Regarding upper-limb OEs, only 2 out of 37 lab and field studies extracted fatigue-related indexes, computed on EMG signals or through indirect physiological measurements, that have been associated with the risk of developing the impingement syndrome or rotator cuff injuries due to repetitive upper-limb tasks (Frost and Andersen, Reference Frost and Andersen1999; Grieve and Dickerson, Reference Grieve and Dickerson2008; Leong et al., Reference Leong, Fu, He, Oh, Yamamoto and Yung2019).

The transfer of laboratory results to the field also depends on the subject group involved. Figure 3 presents the sample sizes from the selected papers. Novice participants or university students are mainly recruited in laboratory studies, whereas field studies include industrial workers, who have a routine in industrial tasks. In laboratory studies, the expert to nonexpert ratio is about 1:3.7 and 1:1.7 in the case of back-support and upper-limb exoskeletons, respectively. Furthermore, the average sample size is higher in laboratory studies than in field studies in the case of upper-limb exoskeletons a ratio of 1.7:1, whereas in the case of back-support exoskeletons the ratio is more balanced (1:1). Exoskeleton evaluations mainly target male participants whereas female end-users have been considered much less frequently.

Figure 3. Sample size analysis of the studies using back-support and upper-limb exoskeletons.

Besides the physical impact of exoskeletons on the human body, their social impact is of major importance for the successful implementation at the workplace. A recent paper by Elprama et al. (Reference Elprama, Vannieuwenhuyze, De Bock, Vanderborght, De Pauw, Meeusen and Jacobs2020) investigated the workers’ intention to use OEs (n = 124). A major gap in literature still exists regarding social measures, such as attitude toward exoskeletons and social perception of exoskeletons. Companies will only implement OEs on a mass scale if they effectively reduce physical and mental loads, and are socially accepted at individual and group levels. Furthermore, acute effects are abundantly researched, but no information is available on the long-term impact of exoskeleton use. Postmarket follow-up studies also allow for an investigation into the cost–benefit and cost-effectiveness of the initial investment.

As visualized in Figure 4, a growing number of papers have been published in recent years that evaluate exoskeletons in laboratory conditions. Over 80% of the reviewed papers describe laboratory studies of back-support (n = 57) and upper-limb exoskeletons (n = 29). Minimum viable products and minimum marketable products are progressively being tested in field situations. Although around 15 exoskeletons are commercially available, only 12 of the identified papers provided insights into field evaluations of back-support (n = 5) and upper-limb exoskeletons (n = 8).

Figure 4. Number of papers that were published in the field of exoskeleton evaluations in laboratory and field conditions.

Roadmap Toward Large-Scale Adoption of OEs

Involving All Stakeholders

Despite the uptake in different domains of application, OEs are still in an early adoption stage. Although certain domains are ahead of others, with the manufacturing industry currently demonstrating the highest uptake, OEs still have not undergone a significant deployment. In this process, a major factor is that several stakeholders may have a different understanding of the costs (e.g., purchase costs and costs related to the loss of productivity due to training sessions) and benefits (e.g., increased occupational health of the workforce and increased productivity due to reduced fatigue and injuries) associated with the adoption of OEs, and may therefore have different opinions on whether and how the OEs can support their specific interests and goals.

As different stakeholders have different interests, they may need different types of information to build their opinion on OEs. Table 2 lists important stakeholders on the demand-side of OEs and indicates their specific goals and interests, that supposedly can be served by this technology, and what type of information they may use to develop their understanding and perspective. Obviously, also stakeholders on the supply side of OEs (such as manufacturers, technology developers, and researchers) will play an important role in creating this market, but, at this stage, it is expected that further advancements of the technology will mainly serve the fulfillment of the needs emerged from the end-users side. It is worth noticing that the information reported in Table 2 has been agreed upon by consensus between the authors and could be further refined by discussing with the stakeholders, with the objective to better define and prioritize the contents.

Table 2. List of stakeholders on the demand side involved in the process of adoption of occupational exoskeletons (OEs)

Note: For each stakeholder, the key interests, goals, and information needs are listed. The information reported in this table has been agreed upon by consensus between the authors.

The list of stakeholders reported in Table 2 highlights the need to build a comprehensive scientific body of knowledge on the effects of OEs on occupational health (from their impact on fatigue and injury to occupational health). This need requires field studies and a variety of measurements spanning from biomechanics and performance, to usability, acceptability, and user experience. Such a holistic assessment would provide evidence of the benefits as well as possible undesired effects of OEs on human health, which would be the first fundamental step to designing novel regulated practices, standards, or methods for ergonomic risk assessment. It is important to note that prevention of health issues typically is a combination of private and public interests, which is reflected in the existing practice around ergonomic and safety-related tools and procedures. Often, guidelines that prescribe what is needed or required in specific sites are based on nonconclusive evidence, through experience-based guidelines, such as the NIOSH lifting equation, or the OCRA-checklist. Such existing frameworks do not yet incorporate a reflection on how OEs interact with such guidelines. In this regard, the release of a new guideline of the Ergonomic Assessment Work-Sheet (EAWS), the EXO-EAWS,Footnote 2 appears to represent an important milestone in the field since this method considers the possibility that the risk assessment index changes with the use of a certified occupational exoskeleton for the first time. Similarly, Di Natali et al. (Reference Di Natali, Chini, Toxiri, Monica, Anastasi, Draicchio, Caldwell and Ortiz2021) proposed a new method to quantify the ergonomic benefit introduced by using a low-back exoskeleton in lifting tasks.

Based on the above considerations, Figure 5 represents a proposed roadmap envisioned by the authors toward the large-scale adoption of OEs. While in the past years, most of the scientific literature has focused on laboratory studies to verify the main biomechanics effects of OEs on the human body (as depicted in S ection “Scientific Evidence of OE Effectiveness in Laboratory and Field Studies”), we believe that in the future, scientific research should more consistently and considerably focus on bringing OEs out of laboratories and testing them in field scenarios. By capturing evidence of effectiveness in several short-term field studies and by using biomechanical measures for the assessment of risk factors for injury, it will become possible to collect success stories from various use cases, which could be the starting point for designing larger-scale studies that can help generalize the findings collected in a specific workplace to a broader application scenario. While, without exoskeletons, large field studies have been carried out to assess the risk of injury at workplaces (Norman et al., Reference Norman, Wells, Neumann, Frank, Shannon and Kerr1998), the state-of-the-art of methodological aspects of evaluating OEs reported in Section “Scientific Evidence of OE Effectiveness in Laboratory and Field Studies” highlights the lack of large-scale studies, as sample sizes are in most cases limited and the generalizability of the results is difficult in almost all application scenarios.

Figure 5. Roadmap toward large-scale adoption of OEs.

Notably, various multinational companies, especially in the automotive field, are currently running large OEs evaluation tests within their plants. However, despite very limited communications at conferences and events, the testing methodology and results of those studies with OEs do not appear to have been presented in scientific publications yet. While it may be possible that companies have limited interest in sharing their data with the scientific community or competitors for strategic reasons, in our opinion the cooperation between companies and the academic/scientific community is paramount for building a solid body of scientific literature on OEs that can lead to the large-scale adoption of this technology in various application scenarios. In addition, workplace insurance boards may play an important role to pursue this type of study, having both the funding and incentive to assess whether the technology can impact the risk of injury, similar to large-scale studies carried out (without OEs) to assess the risk of injury at the workplace (Norman et al., Reference Norman, Wells, Neumann, Frank, Shannon and Kerr1998).

Finally, the output of large field studies may support the development of clear indications about the regulatory context for OEs. Currently, OEs could be classified as medical or nonmedical devices, but a consensus on which choice is more suitable has not been reached yet by the community. In principle, this depends on the intended application domain and claims used on performance and impact of the product, and therefore remains a choice of the manufacturer; nevertheless, this choice is also directed by the expectations of other stakeholders. Some commercial passive OEs have been certified as class I medical devices, according to the European Directive 93/42/EEC, whereas others have been certified as work equipment according to the European directive 2009/104/EC. To the authors’ knowledge, no commercial devices have been certified according to the European Regulation 2016/425 (i.e., the directive on personal protective equipment) to date, but the use of OEs as personal protective equipment has been proposed on several occasions and some attempts have been made to investigate the matter in the United States (Butler and Gillette, Reference Butler and Gillette2019). The different approaches followed by OE manufactures reflect the lack of clear indications of which specific requirements an OE should meet for safe and effective implementation.

Building Evidence Through Considering Use Cases

Although relatively few field studies of OEs have been performed to date, a progressive increase in their number has been noted (Figure 5). However, the variety of methodological approaches used for OE testing (in terms of workstations, experimental methods, metrics, and assessment of risk factors for injury) makes the findings difficult to compare and summarize. In the authors’ opinion, the decision of adopting OEs within a plant requires a stepwise knowledge-based approach, from the observation of the workers’ tasks and routines (to identify the technology that most suits their needs), to the familiarization with and use of the technology (for the assessment of the users’ experience), and up to the final reorganization of the plant space and daily routine to support the adoption of the technology (Figure 5; Masood et al., Reference Masood, Dacal-Nieto, Alonso-Ramos, Fontano, Voilqué and Bou2019; Kerangueven and Atain-Kouadio, Reference Kerangueven, Atain-Kouadio and Turpin-Legendre2020). First of all, understanding the needs of the workers in each specific workplace is a complex yet necessary process for a successful deployment of technology. Aiming at user acceptance in daily practice, identifying the workers’ needs requires use-case-specific effectiveness metrics (Sylla et al., Reference Sylla, Bonnet, Colledani and Fraisse2014; Otten et al., Reference Otten, Weidner and Argubi-Wollesen2018; Torricelli et al., Reference Torricelli, Rodriguez-Guerrero, Veneman, Crea, Briem, Lenggenhager and Beckerle2020). In terms of user-centered design approaches, these could be used to develop customized technology and to evaluate competing commercially available solutions to find the most promising for each use case. Observing the workers’ tasks in their daily practice is paramount to collecting information on which repetitive and nonrepetitive actions are performed (Sylla et al., Reference Sylla, Bonnet, Colledani and Fraisse2014). This facilitates collecting the users’ opinions on their perceived physical effort required to perform strenuous or repetitive actions, which might later be complemented by reports about their experience of the exoskeleton (Shore et al., Reference Shore, de Eyto and O’Sullivan2020).

Furthermore, analyzing the interaction of a worker with the environment is of high importance: several workers might be confined to work in narrow spaces (such as in-car interior assembly), very close to colleagues (such as during handling, cutting, or storing food pieces), or in complex or unstructured environments (such as outdoor gardening or painting). Context-specific information can be complemented by ergonomics risk assessment with well-recognized methods such as OCRA (Occhipinti, Reference Occhipinti1998), EAWS (Schaub et al., Reference Schaub, Mühlstedt, Illmann, Bauer, Fritzsche, Wagner, Bullinger-Hoffmann and Bruder2012), or NIOSH (Dempsey, Reference Dempsey2002). This helps to quantitatively evaluate the effect of repetitive or strenuous actions, and to assess the risk level of the workstation or task under consideration, which could complement the historical data of the companies on the correlation between the development of work-related MSDs and specific workstations.

Analyzing the users’ routines and interaction with the environment, and quantifying ergonomics risks constitutes the basis for evaluating apriori if an OE would have a significant positive impact on the worker’s job quality and health. Such an apriori assessment of the workstation is crucial, as exoskeletons use should only be recommended when indications of the real advantages are clear before any experimentation is performed. This would spare the efforts of testing OEs for situations that do not allow successful exoskeleton implementation, for example, when only a small portion of the time is spent performing the supported movement, or when the movements are highly variable and complex, or when the environment does not allow for a safe introduction of additional tools.

After the deployment, the assessment of OE effectiveness requires, again, a holistic approach (Howard et al., Reference Howard, Murashov, Lowe and Lu2020). While a lot of attention has been put on assessing biomechanical effects, particularly for the target movement, effectiveness seems to be distinctly influenced by psychological and usability factors. Analyzing these calls for further metrics to assess the worker’s experience (Beckerle et al., Reference Beckerle, Salvietti, Unal, Prattichizzo, Rossi, Castellini and Bianchi2017; Torricelli et al., Reference Torricelli, Rodriguez-Guerrero, Veneman, Crea, Briem, Lenggenhager and Beckerle2020), using psychometric techniques, for example, questionnaires (Hart and Staveland, Reference Hart and Staveland1988; Minge et al., Reference Minge, Thüring, Wagner, Kuhr, Soares, Falcão and Ahram2017) or psychophysiological methods, for example, pupil size, heart rate, and skin conductivity (Armel and Ramachandran, Reference Armel and Ramachandran2003; Schultheis and Jameson, Reference Schultheis and Jameson2004; Ikehara and Crosby, Reference Ikehara and Crosby2005). Separate to biomechanical metrics, usability and acceptance can capture aspects that relate to the environment and the organization of the work related to the use of an OE (such as indications for daily practice of effective wearing, storing, and sanitization activities).

Given the above considerations on pre- and post-deployment analyses, the results of the effectiveness evaluation of an OE appear very use-case-specific. Hence, a debate within the scientific community could make it possible to generalize the results of an OE evaluation for specific use-cases to similar job tasks in diverse conditions. Considering that exoskeleton technology is not yet fully mature, many early adopters are cautiously evaluating the OE effectiveness at their specific tasks, which hampers answering this open question. Testing OEs with rigorous methods will certainly help collect information on suitable jobs where the apriori analysis, users’ feedback, and cost-effectiveness agree on their value.

If the testing phase leads to positive results, the decision of the company to adopt this technology requires relatively simple yet necessary modifications of spaces and daily practices to ensure that workers are properly instructed on how to use, store and sanitize the device (similar to personal protective equipment). The reorganization of spaces involves installing dedicated changing rooms or exoskeletons hangers to store the devices. From the perspective of daily routines, the organization of periodic training courses is necessary to instruct and monitor the workers on the correct use, storage, sanitation, and maintenance of the devices. Also, clear indications of when and how to use the exoskeletons during the shift should be provided for the workers and considered in the method-time-measurement analysis of the production line. The decision to introduce exoskeletons into a production line necessarily involves several departments of the company, for example, the health and safety, human resources, and production departments, among others. In medium and large companies, the involvement of unions in the decision process would also ensure that the workers understand this change as an improvement of their health conditions at work.

Finally, in the authors’ opinion, large-scale and longer-term field studies should complement detailed laboratory studies to monitor the effects of daily use of exoskeletons, and particularly to verify the causal relationship between the reduced physical loading of the body and the occurrence of specific work-related MSDs. Such studies can help identify, monitor, and quantify the possible undesired side-effects related to exoskeleton use (Howard et al., Reference Howard, Murashov, Lowe and Lu2020), as well as provide solid scientific knowledge to support the revision of ergonomics risk-assessment methods, the definition of clear guidelines for selecting OEs (such as quality marks to distinguish between good or bad technologies), practices to use them, and the revision of the related safety standards and regulations.

Conclusion

Considering that OEs are still in an early adoption stage, large-scale deployment of this technology will necessarily go through a more structured involvement of all stakeholders, such as the workers, the health and safety, human resources and production departments, as well as the unions and the policymakers. All stakeholders should be informed of the costs and benefits of OEs from their perspectives and interests, as they can have different objectives and therefore require different knowledge to make their decisions. From this context, it is important to consider larger field studies not just as scientific studies, but also as experiments-in-practice, that allow all stakeholders to determine and develop their understanding. It is therefore of key importance that all relevant stakeholders are to some degree involved in the definition of any larger studies in the field, in order to maximize the impact of the studies on an eventual larger uptake of OEs.

The large-scale adoption of OEs will require a stepwise knowledge-based approach, grounded on careful investigation of work task biomechanics, usability, acceptability, and user experience. Thus, we believe that in the future, scientific research will need to consistently focus on complementing laboratory studies with field investigations. By capturing evidence of effectiveness in several use cases, through short-term field studies, it will become possible to start designing larger-scale studies. Such studies are of paramount importance since they could help to generalize the findings collected in specific workplaces to broader application scenarios. Finally, longer-term field studies should be carried out with the key objective to verify whether OEs can reduce the occurrence of specific work-related MSDs.

Acknowledgments

This article is based upon work from COST Action 16116—Wearable Robots for Augmentation, Assistance or Substitution of Human Motor Functions, supported by COST (European Cooperation in Science and Technology).

Supplementary Materials

To view supplementary material for this article, please visit http://dx.doi.org/10.1017/wtc.2021.11.

Authorship Contributions

All authors contributed to conceptualize the work and interpret the data. S.C. and P.B. drafted the section “Introduction.” S.C., L.G., I.P., and C.R.-G. drafted the section “Technological Approaches to Design OEs.” M.D.L., K.D.P., L.G., T.K., J.M., L.W.O., I.P., and D.R.-D. drafted the section “Scientific Evidence of OE Effectiveness in Laboratory and Field Studies.” N.V. and J.V. drafted the section “Roadmap Toward Large-Scale Adoption of OEs.”

Funding Statement

This research has received funding from the COST Action CA16116—“Wearable Robots for Augmentation, Assistance or Substitution of Human Motor Functions.” K.D.P. is funded by Strategic Research Program “Exercise and the Brain in Health & Disease: The Added Value of Human-Centered Robotics,” Vrije Universiteit Brussel (Belgium). The contributions by T.K. and L.O. are supported by research funded by Science Foundation Ireland (SFI) under Grant Number SFI 16/RC/3918. J.M. research contribution is supported by F4.0 Automoción and respectively to the agency GAIN from the Xunta de Galicia regional government of Spain.

Competing Interests

S.C. and N.V. have interests in IUVO S.r.l., a spin-off company of Scuola Superiore Sant’Anna, which develops wearable robots. The remaining authors declare no competing interests exist.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Abbreviations

EMG

electromyography

OE

occupational exoskeleton

RPE

rate of perceived exertion

SUS

system usability scale

WRMSDs

work-related musculoskeletal disorders

References

Abdoli-E, M, Agnew, MJ and Stevenson, JM (2006) An on-body personal lift augmentation device (PLAD) reduces EMG amplitude of erector spinae during lifting tasks. Clinical Biomechanics 21(5), 456465. https://doi.org/10.1016/j.clinbiomech.2005.12.021CrossRefGoogle ScholarPubMed
Accoto, D, Sergi, F, Tagliamonte, NL, Carpino, G, Sudano, A and Guglielmelli, E (2014) Robomorphism: A nonanthropomorphic wearable robot. IEEE Robotics and Automation Magazine 21(4), 4555. https://doi.org/10.1109/MRA.2014.2360276CrossRefGoogle Scholar
Alabdulkarim, S and Nussbaum, MA (2019) Influences of different exoskeleton designs and tool mass on physical demands and performance in a simulated overhead drilling task. Applied Ergonomics 74, 5566. https://doi.org/10.1016/j.apergo.2018.08.004CrossRefGoogle Scholar
Alemi, MM, Geissinger, J, Simon, AA, Chang, SE and Asbeck, AT (2019) A passive exoskeleton reduces peak and mean EMG during symmetric and asymmetric lifting. Journal of Electromyography and Kinesiology 47, 2534. https://doi.org/10.1016/j.jelekin.2019.05.003CrossRefGoogle ScholarPubMed
Amandels, S, het Eyndt, HO, Daenen, L and Hermans, V (2019) Introduction and testing of a passive exoskeleton in an industrial working environment. In Advances in Intelligent Systems and Computing (Vol. 820). Cham: Springer International Publishing. http://dx.doi.org/10.1007/978-3-319-96083-8_51Google Scholar
Anton, D, Shibley, LD, Fethke, NB, Hess, J, Cook, TM and Rosecrance, J (2001) The effect of overhead drilling position on shoulder moment and electromyography. Ergonomics 44(5), 489501. https://doi.org/10.1080/00140130120079CrossRefGoogle ScholarPubMed
Ármannsdóttir, AL, Beckerle, P, Moreno, JC, van Asseldonk, EHF, Manrique-Sancho, MT, Del-Ama, AJ, Veneman, JF and Briem, K (2020) Assessing the involvement of users during development of lower limb wearable robotic exoskeletons: A survey Study. Human Factors 62(3), 351364.CrossRefGoogle ScholarPubMed
Armel, KC and Ramachandran, VS (2003) Projecting sensations to external objects: Evidence from skin conductance response. Proceedings of the Royal Society of London. Series B: Biological Sciences 270(1523), 14991506. https://doi.org/10.1098/rspb.2003.2364CrossRefGoogle ScholarPubMed
Baltrusch, SJ, Houdijk, H, van Dieën, JH and Kruif, JTCM (2020a) Passive trunk exoskeleton acceptability and effects on self-efficacy in employees with low-back pain: A mixed method approach. Journal of Occupational Rehabilitation 31, 129141. http://dx.doi.org/10.1007/s10926-020-09891-1CrossRefGoogle Scholar
Baltrusch, SJ, van Dieën, JH, Bruijn, SM, Koopman, AS, van Bennekom, CAM and Houdijk, H (2019) The effect of a passive trunk exoskeleton on metabolic costs during lifting and walking. Ergonomics 62(7), 903916. https://doi.org/10.1080/00140139.2019.1602288CrossRefGoogle ScholarPubMed
Baltrusch, SJ, van Dieën, JH, Koopman, AS, Näf, MB, Rodriguez-Guerrero, C, Babič, J and Houdijk, H (2020b) SPEXOR passive spinal exoskeleton decreases metabolic cost during symmetric repetitive lifting. European Journal of Applied Physiology 120(2), 401412. https://doi.org/10.1007/s00421-019-04284-6CrossRefGoogle Scholar
Baltrusch, SJ, van Dieen, JH, van Bennekom, CAM and Houdijk, H (2020c) Testing an exoskeleton that helps workers with low-back pain: Less discomfort with the passive SPEXOR trunk device. IEEE Robotics & Automation Magazine 27(1), 6676. https://doi.org/10.1109/MRA.2019.2954160CrossRefGoogle Scholar
Baltrusch, SJ, van Dieën, JH, van Bennekom, CAM and Houdijk, H (2018) The effect of a passive trunk exoskeleton on functional performance in healthy individuals. Applied Ergonomics 72, 94106. https://doi.org/10.1016/j.apergo.2018.04.007CrossRefGoogle ScholarPubMed
Basmajian, JV and De Luca, CJ (1985) Muscles Alive: Their Functions Revealed by Electromyography, 5th Edn. Baltimore, MD: Williams & Wilkins.Google Scholar
Beckerle, P, Salvietti, G, Unal, R, Prattichizzo, D, Rossi, S, Castellini, C and Bianchi, M (2017) A human–robot interaction perspective on assistive and rehabilitation robotics. Frontiers in Neurorobotics 11, 24. https://doi.org/10.3389/fnbot.2017.00024CrossRefGoogle ScholarPubMed
Blanco, A, Catalán, JM, Díez, JA, García, JV, Lobato, E and García-Aracil, N (2019) Electromyography assessment of the assistance provided by an upper-limb exoskeleton in maintenance tasks. Sensors 19(5), 3391. https://doi.org/10.3390/s19153391CrossRefGoogle ScholarPubMed
Blanco, A, Martinez, D, Catalán, JM, García, JV, Ezquerro, S, Díez, JA and García-Aracil, N (2020) Oxygen consumption in industrial tasks assisted by an active upper-limb exoskeleton. In Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics. Singapore: IEEE. http://dx.doi.org/10.1109/BioRob49111.2020.9224385Google Scholar
Borg, GAV (1982) Psychophysical bases of perceived exertion. Medicine & Science in Sports & Exercise 14(5), 377381.CrossRefGoogle Scholar
Bosch, T, van Eck, J, Knitel, K and de Looze, M (2016) The effects of a passive exoskeleton on muscle activity, discomfort and endurance time in forward bending work. Applied Ergonomics 54, 212217. https://doi.org/10.1016/j.apergo.2015.12.003CrossRefGoogle ScholarPubMed
Butler, T and Gillette, JC (2019) Exoskeletons: Used as PPE for injury prevention. Professional Safety Journal 3, 3237.Google Scholar
Chen, B, Grazi, L, Lanotte, F, Vitiello, N and Crea, S (2018) A real-time lift detection strategy for a hip exoskeleton. Frontiers in Neurorobotics 12, 17. https://doi.org/10.3389/fnbot.2018.00017CrossRefGoogle ScholarPubMed
Claramunt-Molet, M, Domingo-Mateu, B, Danús-Jaume, J, Ugartemendia-Etxarri, A, De La Maza, I, Enriquez-Carrera, V, Muñoz-Fernandez, O, Domingo, S, Miralles, F, Font-Llagunes, JM and Idelsohn-Zielonka, S (2019) Biomechanical evaluation of upper limb exoskeletons in automotive assembly using EMG. In Reunión Del Capítulo Español de La Sociedad Europea de Biomecánica. Barcelona: Grupo de Investigación de Biomateriales y Biomecánica, pp. 2526.Google Scholar
De Bock, S, Ghillebert, J, Govaerts, R, Elprama, SA, Marusic, U, Serrien, B, Jacobs, A, Geeroms, J, Meeusen, R and De Pauw, K (2020) Passive shoulder exoskeletons: More effective in the lab than in the field? IEEE Transactions on Neural Systems and Rehabilitation Engineering 29(2016), 173183. https://doi.org/10.1109/TNSRE.2020.3041906CrossRefGoogle Scholar
de Looze, MP, Bosch, T, Krause, F, Stadler, KS and O’Sullivan, LW (2016) Exoskeletons for industrial application and their potential effects on physical work load. Ergonomics 59, 671681. http://dx.doi.org/10.1080/00140139.2015.1081988CrossRefGoogle ScholarPubMed
de Vries, A and de Looze, M (2020) The effect of arm support exoskeletons in realistic work activities: A review Study. Journal of Ergonomics 9(4), 19. https://doi.org/10.35248/2165-7556.19.9.255Google Scholar
de Vries, AW, Krause, F and de Looze, MP (2021) The effectivity of a passive arm support exoskeleton in reducing muscle activation and perceived exertion during plastering activities. Ergonomics 64, 712721. http://dx.doi.org/10.1080/00140139.2020.1868581CrossRefGoogle ScholarPubMed
de Vries, A, Murphy, M, Könemann, R, Kingma, I and de Looze, M (2019) The amount of support provided by a passive arm support exoskeleton in a range of elevated arm postures. IISE Transactions on Occupational Ergonomics and Human Factors 7(3–4), 311321. https://doi.org/10.1080/24725838.2019.1669736CrossRefGoogle Scholar
Dempsey, PG (2002) Usability of the revised NIOSH lifting equation. Ergonomics 45(12), 817828.CrossRefGoogle ScholarPubMed
Di Natali, C, Chini, G, Toxiri, S, Monica, L, Anastasi, S, Draicchio, F, Caldwell, DG and Ortiz, J (2021) Equivalent weight: Connecting exoskeleton effectiveness with ergonomic risk during manual material handling. International Journal of Environmental Research and Public Health 18(5), 2677. https://doi.org/10.3390/ijerph18052677CrossRefGoogle ScholarPubMed
Elprama, SA, Vannieuwenhuyze, JTA, De Bock, S, Vanderborght, B, De Pauw, K, Meeusen, R and Jacobs, A (2020) Social processes: What determines industrial workers. Intention to Use Exoskeletons? Human Factors 62(3), 337350. https://doi.org/10.1177/0018720819889534Google Scholar
Ferreira, G, Gaspar, J, Fujão, C and Nunes, IL (2020) Piloting the use of an upper limb passive exoskeleton in automotive industry: Assessing user acceptance and intention of use BT. In Nunes, IL (ed.), Advances in Human Factors and Systems Interaction. Cham: Springer International Publishing, pp. 342349. http://dx.doi.org/10.1007/978-3-030-51369-6_46CrossRefGoogle Scholar
Fox, S, Aranko, O, Heilala, J and Vahala, P (2019) Exoskeletons: Comprehensive, comparative and critical analyses of their potential to improve manufacturing performance. Journal of Manufacturing Technology Management 31(6), 12611280. https://doi.org/10.1108/JMTM-01-2019-0023CrossRefGoogle Scholar
Frost, P and Andersen, JH (1999) Shoulder impingement syndrome in relation to shoulder intensive work. Occupational and Environmental Medicine 56(7), 494498. https://doi.org/10.1136/oem.56.7.494CrossRefGoogle ScholarPubMed
Gillette, JC and Stephenson, ML (2018) EMG analysis of an upper body exoskeleton during automotive assembly. Proceedings of the 42nd Annual Meeting of the American Society of Biomechanics, August, pp. 308–309.Google Scholar
Gillette, JC and Stephenson, ML (2019) Electromyographic assessment of a shoulder support exoskeleton during on-site job tasks. IISE Transactions on Occupational Ergonomics and Human Factors 7(3–4), 302310. https://doi.org/10.1080/24725838.2019.1665596CrossRefGoogle Scholar
Goršič, M, Regmi, Y, Johnson, AP, Dai, B and Novak, D (2019) A pilot study of varying thoracic and abdominal compression in a reconfigurable trunk exoskeleton during different activities. IEEE Transactions on Biomedical Engineering 67(6), 1. https://doi.org/10.1109/TBME.2019.2940431Google Scholar
Graham, RB, Agnew, MJ and Stevenson, JM (2009) Effectiveness of an on-body lifting aid at reducing low back physical demands during an automotive assembly task: Assessment of EMG response and user acceptability. Applied Ergonomics 40(5), 936942. https://doi.org/10.1016/j.apergo.2009.01.006CrossRefGoogle ScholarPubMed
Grazi, L, Trigili, E, Proface, G, Giovacchini, F, Crea, S and Vitiello, N (2020) Design and experimental evaluation of a semi-passive upper-limb exoskeleton for workers with motorized tuning of assistance. IEEE Transactions on Neural Systems and Rehabilitation Engineering 28, 22762285. http://dx.doi.org/10.1109/tnsre.2020.3014408CrossRefGoogle ScholarPubMed
Grieve, JR and Dickerson, CR (2008) Overhead work: Identification of evidence-based exposure guidelines. Occupational Ergonomics 8(1), 5366. https://doi.org/10.3233/OER-2008-8105CrossRefGoogle Scholar
Gull, MA, Bai, S and Bak, T (2020) A review on design of upper limb exoskeletons. Robotics 9(1), 16. https://doi.org/10.3390/robotics9010016CrossRefGoogle Scholar
Hart, SG and Staveland, LE (1988) Development of NASA-TLX (task load index): Results of empirical and theoretical research. Advances in Psychology 52, 139183.CrossRefGoogle Scholar
Hefferle, M, Snell, M and Kluth, K (2020) Influence of two industrial overhead exoskeletons on perceived strain – A field study in the automotive industry. Paper Presented at the International Conference on Applied Human Factors and Ergonomics.CrossRefGoogle Scholar
Hensel, R and Keil, M (2019) Subjective evaluation of a passive industrial exoskeleton for lower-back support: A field study in the automotive sector. IISE Transactions on Occupational Ergonomics and Human Factors 7(3–4), 213221. https://doi.org/10.1080/24725838.2019.1573770CrossRefGoogle Scholar
Heo, U, Kim, SJ and Kim, J (2020) Backdrivable and fully-portable pneumatic back support exoskeleton for lifting assistance. IEEE Robotics and Automation Letters 5(2), 20472053. https://doi.org/10.1109/LRA.2020.2969169CrossRefGoogle Scholar
Howard, J, Murashov, VV, Lowe, BD and Lu, ML (2020) Industrial exoskeletons: Need for intervention effectiveness research. American Journal of Industrial Medicine 63(3), 201208. https://doi.org/10.1002/ajim.23080CrossRefGoogle ScholarPubMed
Hussain, M, Park, J, Kim, N, Kim, HK and Lee, J (2020) Effects of exoskeleton robot on human posture and lumbar pressure during manual lifting tasks. ICIC Express Letters, Part B: Applications 11(5), 439445. https://doi.org/10.24507/icicelb.11.05.439Google Scholar
Huysamen, K, Bosch, T, de Looze, M, Stadler, KS, Graf, E and O’Sullivan, LW (2018a) Evaluation of a passive exoskeleton for static upper limb activities. Applied Ergonomics 70, 148155. https://doi.org/10.1016/j.apergo.2018.02.009CrossRefGoogle Scholar
Huysamen, K, de Looze, MP, Bosch, T, Ortiz, J, Toxiri, S and O’Sullivan, L (2018b) Assessment of an active industrial exoskeleton to aid dynamic lifting and lowering manual handling tasks. Applied Ergonomics 68, 125131. http://dx.doi.org/10.1016/j.apergo.2017.11.004CrossRefGoogle Scholar
Huysamen, K, Power, V and O’Sullivan, L (2020) Kinematic and kinetic functional requirements for industrial exoskeletons for lifting tasks and overhead lifting. Ergonomics 63(7), 818830. https://doi.org/10.1080/00140139.2020.1759698CrossRefGoogle ScholarPubMed
Hyun, DJ, Lim, H, Park, S and Nam, S (2020) Singular wire-driven series elastic actuation with force control for a waist assistive exoskeleton, H-WEXv2. IEEE/ASME Transactions on Mechatronics 25(2), 10261035. https://doi.org/10.1109/TMECH.2020.2970448CrossRefGoogle Scholar
Ikehara, CS and Crosby, ME (2005) Assessing cognitive load with physiological sensors. In Proceedings of the 38th Annual Hawaii International Conference on System Sciences. Big Island, HI: IEEE, p. 295. http://dx.doi.org/10.1109/HICSS.2005.103CrossRefGoogle Scholar
Inose, H, Mohri, S, Arakawa, H, Okui, M, Koide, K, Yamada, Y, Kikutani, I and Nakamura, T (2017) Semi-endoskeleton-type waist assist AB-wear suit equipped with compressive force reduction mechanism. In 2017 IEEE International Conference on Robotics and Automation (ICRA). Singapore: IEEE, pp. 60146019. http://dx.doi.org/10.1109/ICRA.2017.7989711CrossRefGoogle Scholar
Iranzo, S, Piedrabuena, A, Iordanov, D, Martinez-Iranzo, U and Belda-Lois, J-M (2020) Ergonomics assessment of passive upper-limb exoskeletons in an automotive assembly plant. Applied Ergonomics 87, 103120. http://dx.doi.org/10.1016/j.apergo.2020.103120CrossRefGoogle Scholar
Jamšek, M, Petrič, T and Babič, J (2020) Gaussian mixture models for control of quasi-passive spinal exoskeletons. Sensors 20(9), 2705. https://doi.org/10.3390/s20092705CrossRefGoogle ScholarPubMed
Kazerooni, H, Tung, W and Pillai, M (2019) Evaluation of trunk-supporting exoskeleton. Proceedings of the Human Factors and Ergonomics Society Annual Meeting 63(1), 10801083. https://doi.org/10.1177/1071181319631261CrossRefGoogle Scholar
Kerangueven, L, Atain-Kouadio, JJ and Turpin-Legendre, E. (2020) Acquisition and integration of exoskeletons in establishments, p. 29.Google Scholar
Kermavnar, T, de Vries, AW, de Looze, MP and O’Sullivan, LW (2020) Effects of industrial back-support exoskeletons on body loading and user experience: An updated systematic review. Ergonomics 64, 685711. http://dx.doi.org/10.1080/00140139.2020.1870162CrossRefGoogle Scholar
Kim, YG, Little, K, Noronha, B, Xiloyannis, M, Masia, L and Accoto, D (2020a) A voice activated bi-articular exosuit for upper limb assistance during lifting tasks. Robotics and Computer-Integrated Manufacturing 66, 101995. http://dx.doi.org/10.1016/j.rcim.2020.101995CrossRefGoogle Scholar
Kim, S, Madinei, S, Alemi, MM, Srinivasan, D and Nussbaum, MA (2020b) Assessing the potential for “undesired” effects of passive back-support exoskeleton use during a simulated manual assembly task: Muscle activity, posture, balance, discomfort, and usability. Applied Ergonomics 89, 103194. https://doi.org/10.1016/j.apergo.2020.103194CrossRefGoogle Scholar
Kim, S, Nussbaum, MA, Mokhlespour Esfahani, MI, Alemi, MM, Alabdulkarim, S and Rashedi, E (2018) Assessing the influence of a passive, upper extremity exoskeletal vest for tasks requiring arm elevation: Part I – “Expected” effects on discomfort, shoulder muscle activity, and work task performance. Applied Ergonomics 70, 315322. https://doi.org/10.1016/j.apergo.2018.02.025CrossRefGoogle ScholarPubMed
Ko, HK, Lee, SW, Koo, DH, Lee, I and Hyun, DJ (2018) Waist-assistive exoskeleton powered by a singular actuation mechanism for prevention of back-injury. Robotics and Autonomous Systems 107, 19. https://doi.org/10.1016/j.robot.2018.05.008CrossRefGoogle Scholar
Koopman, AS, Kingma, I, de Looze, MP and van Dieën, JH (2020b) Effects of a passive back exoskeleton on the mechanical loading of the low-back during symmetric lifting. Journal of Biomechanics 102, 109486. https://doi.org/10.1016/j.jbiomech.2019.109486CrossRefGoogle Scholar
Koopman, AS, Kingma, I, Faber, GS, de Looze, MP and van Dieën, JH (2019a) Effects of a passive exoskeleton on the mechanical loading of the low back in static holding tasks. Journal of Biomechanics 83, 97103. https://doi.org/10.1016/j.jbiomech.2018.11.033CrossRefGoogle Scholar
Koopman, AS, Näf, M, Baltrusch, SJ, Kingma, I, Rodriguez-Guerrero, C, Babič, J, Kingma, I, Rodriguez-Guerrero, C, Babič, J, de Looze, MP and van Dieën, JH (2020a) Biomechanical evaluation of a new passive back support exoskeleton. Journal of Biomechanics 105, 109795. https://doi.org/10.1016/j.jbiomech.2020.109795CrossRefGoogle Scholar
Koopman, AS, Toxiri, S, Power, V, Kingma, I, van Dieën, JH, Ortiz, J and de Looze, MP (2019b) The effect of control strategies for an active back-support exoskeleton on spine loading and kinematics during lifting. Journal of Biomechanics 91, 1422. https://doi.org/10.1016/j.jbiomech.2019.04.044CrossRefGoogle Scholar
Lamers, EP, Yang, AJ and Zelik, KE (2018) Feasibility of a biomechanically-assistive garment to reduce low back loading during leaning and lifting. IEEE Transactions on Biomedical Engineering 65(8), 16741680. https://doi.org/10.1109/TBME.2017.2761455Google ScholarPubMed
Lamers, EP and Zelik, KE (2021) Design, modeling, and demonstration of a new dual-mode back-assist exosuit with extension mechanism. Wearable Technologies 2, 126. https://doi.org/10.1017/wtc.2021.1CrossRefGoogle Scholar
Lanotte, F, Baldoni, A, Dell’Agnello, F, Scalamogna, A, Mansi, N, Grazi, L, Chen, B, Crea, S and Vitiello, N (2020) Design and characterization of a multi-joint underactuated low-back exoskeleton for lifting tasks. In 2020 8th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob). New York: IEEE, pp. 11461151. http://dx.doi.org/10.1109/BioRob49111.2020.9224370CrossRefGoogle Scholar
Lanotte, F, Grazi, L, Chen, B, Vitiello, N and Crea, S (2018) A low-back exoskeleton can reduce the erector spinae muscles activity during freestyle symmetrical load lifting tasks. In 2018 7th IEEE International Conference on Biomedical Robotics and Biomechatronics (Biorob). Enschede, Netherlands: IEEE, pp. 701706. http://dx.doi.org/10.1109/BIOROB.2018.8488094CrossRefGoogle Scholar
Lanotte, F, McKinney, Z, Grazi, L, Chen, B, Crea, S and Vitiello, N (2021) Adaptive control method for dynamic synchronization of wearable robotic assistance to discrete movements: Validation for use case of lifting tasks. IEEE Transactions on Robotics. http://dx.doi.org/10.1109/TRO.2021.3073836CrossRefGoogle Scholar
Leong, HT, Fu, SC, He, X, Oh, JH, Yamamoto, N and Yung, SHP (2019) Risk factors for rotator cuff tendinopathy: A systematic review and meta-analysis. Journal of Rehabilitation Medicine 51(9), 627637. https://doi.org/10.2340/16501977-2598CrossRefGoogle ScholarPubMed
Madinei, S, Alemi, MM, Kim, S, Srinivasan, D and Nussbaum, MA (2020a) Biomechanical assessment of two back-support exoskeletons in symmetric and asymmetric repetitive lifting with moderate postural demands. Applied Ergonomics 88, 103156. https://doi.org/10.1016/j.apergo.2020.103156CrossRefGoogle Scholar
Madinei, S, Alemi, MM, Kim, S, Srinivasan, D and Nussbaum, MA (2020b) Biomechanical evaluation of passive back-support exoskeletons in a precision manual assembly task: “Expected” effects on trunk muscle activity, perceived exertion, and task performance. Human Factors 62(3), 441457. http://dx.doi.org/10.1177/0018720819890966CrossRefGoogle Scholar
Masood, J, Dacal-Nieto, A, Alonso-Ramos, V, Fontano, MI, Voilqué, A and Bou, J (2019) Industrial wearable exoskeletons and exosuits assessment process. Biosystems and Biorobotics 22, 234238. https://doi.org/10.1007/978-3-030-01887-0_45CrossRefGoogle Scholar
Maurice, P, Čamernik, J, Gorjan, D, Schirrmeister, B, Bornmann, J, Tagliapietra, L, Latella, C, Pucci, D, Fritzsche, L, Ivaldi, S and Babič, J (2019) Objective and subjective effects of a passive exoskeleton on overhead work. IEEE Transactions on Neural Systems and Rehabilitation Engineering 28(1), 152164. https://doi.org/10.1109/TNSRE.2019.2945368CrossRefGoogle ScholarPubMed
McFarland, T and Fischer, S (2019) Considerations for industrial use: A systematic review of the impact of active and passive upper limb exoskeletons on physical exposures. IISE Transactions on Occupational Ergonomics and HF 7, 322347. https://doi.org/10.1080/24725838.2019.1684399CrossRefGoogle Scholar
McGill, SM and Kippers, V (1994) Transfer of loads between lumbar tissues during the flexion-relaxation phenomenon. Spine 19(19), 21902196. https://doi.org/10.1097/00007632-199410000-00013CrossRefGoogle ScholarPubMed
Minge, M, Thüring, M, Wagner, I and Kuhr, CV (2017) The meCUE Questionnaire: A modular tool for measuring user experience BT. In Soares, M, Falcão, C and Ahram, TZ (eds), Advances in Ergonomics Modeling, Usability & Special Populations. Cham: Springer International Publishing, pp. 115128. https://doi.org/10.1007/978-3-319-41685-4_11CrossRefGoogle Scholar
Miura, K, Kadone, H, Koda, M, Abe, T, Endo, H, Murakami, H, Doita, M, Kumagai, H, Nagashima, K, Fujii, K, Noguchi, H, Funayama, T, Kawamoto, H, Sankai, Y and Yamazaki, M (2018) The hybrid assisted limb (HAL) for care support, a motion assisting robot providing exoskeletal lumbar support, can potentially reduce lumbar load in repetitive snow-shoveling movements. Journal of Clinical Neuroscience 49, 8386. https://doi.org/10.1016/j.jocn.2017.11.020CrossRefGoogle ScholarPubMed
Monica, L, Anastasi, S and Draicchio, F (2020) Occupational exoskeletons: Wearable robotic devices and preventing work-related musculoskeletal disorders in the workplace of the future, pp. 1–12. Available at https://osha.europa.eu/en/publications/.Google Scholar
Motmans, R, Debaets, T and Chrispeels, S (2019) Effect of a passive exoskeleton on muscle activity and posture during order picking. Advances in Intelligent Systems and Computing 820, 338346. https://doi.org/10.1007/978-3-319-96083-8_45CrossRefGoogle Scholar
Moyon, A, Poirson, E and Petiot, J-F (2018) Experimental study of the physical impact of a passive exoskeleton on manual sanding operations. Procedia CIRP 70, 284289. https://doi.org/10.1016/j.procir.2018.04.028CrossRefGoogle Scholar
Naf, MB, Junius, K, Rossini, M, Rodriguez-Guerrero, C, Vanderborght, B and Lefeber, D (2018) Misalignment compensation for full human-exoskeleton kinematic compatibility: State of the art and evaluation. Applied Mechanics Reviews 70(5), 050802. https://doi.org/10.1115/1.4042523CrossRefGoogle Scholar
Näf, MB, Koopman, AS, Baltrusch, S, Rodriguez-Guerrero, C, Vanderborght, B and Lefeber, D (2018) Passive Back support exoskeleton improves range of motion using flexible beams. Frontiers in Robotics and AI 5, 72. https://doi.org/10.3389/frobt.2018.00072CrossRefGoogle ScholarPubMed
Naito, J, Obinata, G, Nakayama, A and Hase, K (2007) Development of a wearable robot for assisting carpentry workers. International Journal of Advanced Robotic Systems 4(4), 48. https://doi.org/10.5772/5667CrossRefGoogle Scholar
Nef, T, Guidali, M and Riener, R (2009) ARMin III – Arm therapy exoskeleton with an ergonomic shoulder actuation. Applied Bionics and Biomechanics 6, 127142. https://doi.org/10.1080/11762320902840179CrossRefGoogle Scholar
Norman, R, Wells, R, Neumann, P, Frank, J, Shannon, H and Kerr, M (1998) A comparison of peak vs cumulative physical work exposure risk factors for the reporting of low back pain in the automotive industry. Clinical Biomechanics 13(8), 561573. http://dx.doi.org/10.1016/S0268-0033(98)00020-5CrossRefGoogle ScholarPubMed
Occhipinti, E (1998) OCRA: A concise index for the assessment of exposure to repetitive movements of the upper limbs. Ergonomics 41(9), 12901311. https://doi.org/10.1080/001401398186315CrossRefGoogle ScholarPubMed
Otten, BM, Weidner, R and Argubi-Wollesen, A (2018) Evaluation of a novel active exoskeleton for tasks at or above head level. IEEE Robotics and Automation Letters 3(3), 24082415. https://doi.org/10.1109/LRA.2018.2812905CrossRefGoogle Scholar
Pacifico, I, Scano, A, Guanziroli, E, Moise, M, Morelli, L, Chiavenna, A, Romo, D, Spada, S, Colombina, G, Molteni, F, Giovacchini, F, Vitiello, N and Crea, S (2020) An experimental evaluation of the proto-MATE: A novel ergonomic upper-limb exoskeleton to reduce workers’ physical strain. IEEE Robotics and Automation Magazine 27(1), 5465. https://doi.org/10.1109/MRA.2019.2954105CrossRefGoogle Scholar
Perez Luque, E, Högberg, D, Iriondo Pascual, A, Lämkull, D and Garcia Rivera, F (2020) Motion behavior and range of motion when using exoskeletons in manual assembly tasks. SPS2020: Prooceedings of the Swedish Production Symposium 13, 217228. https://doi.org/10.3233/ATDE200159Google Scholar
Picchiotti, MT, Weston, EB, Knapik, GG, Dufour, JS and Marras, WS (2019) Impact of two postural assist exoskeletons on biomechanical loading of the lumbar spine. Applied Ergonomics 75, 17. https://doi.org/10.1016/j.apergo.2018.09.006CrossRefGoogle ScholarPubMed
Poliero, T, Lazzaroni, M, Toxiri, S, Di Natali, C, Caldwell, DG and Ortiz, J (2020) Applicability of an active Back-support exoskeleton to carrying activities. Frontiers in Robotics and AI 7, 579963. https://doi.org/10.3389/frobt.2020.579963CrossRefGoogle ScholarPubMed
Rashedi, E, Kim, S, Nussbaum, MA and Agnew, MJ (2014) Ergonomic evaluation of a wearable assistive device for overhead work. Ergonomics 57(12), 18641874. https://doi.org/10.1080/00140139.2014.952682CrossRefGoogle ScholarPubMed
Rossini, M, Geeroms, J, Lefeber, D and Rodriguez-Guerrero, C (2021) Automatic synthesis of arthrokinematically compatible exoskeletons. A case study on its application on a shoulder occupational exoskeleton. Mechanism and Machine Theory 157, 104186. https://doi.org/10.1016/j.mechmachtheory.2020.104186CrossRefGoogle Scholar
Saraceni, N, Campbell, A, Kent, P, Ng, L, Straker, L and O’Sullivan, P (2021) Exploring lumbar and lower limb kinematics and kinetics for evidence that lifting technique is associated with LBP. PLoS One 16, 120. http://dx.doi.org/10.1371/journal.pone.0254241CrossRefGoogle ScholarPubMed
Schaub, KG, Mühlstedt, J, Illmann, B, Bauer, S, Fritzsche, L, Wagner, T, Bullinger-Hoffmann, AC and Bruder, R (2012) Ergonomic assessment of automotive assembly tasks with digital human modelling and the “ergonomics assessment worksheet” (EAWS). International Journal of Human Factors Modelling and Simulation 3(3/4), 398. https://doi.org/10.1504/ijhfms.2012.051581CrossRefGoogle Scholar
Schiele, A and van der Helm, FCT (2006) Kinematic design to improve ergonomics in human machine interaction. IEEE Transactions on Neural Systems and Rehabilitation Engineering 14(4), 456469. https://doi.org/10.1109/TNSRE.2006.881565CrossRefGoogle ScholarPubMed
Schultheis, H and Jameson, A (2004) Assessing cognitive load in adaptive hypermedia systems: Physiological and behavioral methods. In International Conference on Adaptive Hypermedia and Adaptive Web-Based Systems. Berlin: Springer, pp. 225234.CrossRefGoogle Scholar
Schmalz, T, Schändlinger, J, Schuler, M, Bornmann, J, Schirrmeister, B, Kannenberg, A and Ernst, M (2019) Biomechanical and Metabolic Effectiveness of an Industrial Exoskeleton for Overhead Work. International Journal of Environmental Research and Public Health, 16(23), 4792. https://doi.org:10.3390/ijerph16234792CrossRefGoogle ScholarPubMed
Settembre, N, Maurice, P, Paysant, J, Theurel, J, Claudon, L, Kimmoun, A, Levy, B, Hani, H, Chenuel, B and Ivaldi, S (2020) The use of exoskeletons to help with prone positioning in the intensive care unit during COVID-19. Annals of Physical and Rehabilitation Medicine 63(4), 379382. https://doi.org/10.1016/j.rehab.2020.05.004CrossRefGoogle ScholarPubMed
Shore, L, de Eyto, A and O’Sullivan, L (2020) Technology acceptance and perceptions of robotic assistive devices by older adults – implications for exoskeleton design. Disability and Rehabilitation: Assistive Technology, 19. http://dx.doi.org/10.1080/17483107.2020.1817988Google ScholarPubMed
Smets, M (2019) A field evaluation of arm-support exoskeletons for overhead work applications in automotive assembly. IISE Transactions on Occupational Ergonomics and Human Factors 7(3–4), 192198. https://doi.org/10.1080/24725838.2018.1563010CrossRefGoogle Scholar
Spada, S, Ghibaudo, L, Gilotta, S, Gastaldi, L and Cavatorta, MP (2017) Investigation into the applicability of a passive upper-limb exoskeleton in automotive industry. Procedia Manufacturing 11, 12551262. https://doi.org/10.1016/j.promfg.2017.07.252CrossRefGoogle Scholar
Sylla, N, Bonnet, V, Colledani, F and Fraisse, P (2014) Ergonomic contribution of ABLE exoskeleton in automotive industry. International Journal of Industrial Ergonomics 44(4), 475481. https://doi.org/10.1016/j.ergon.2014.03.008CrossRefGoogle Scholar
Tan, CK, Kadone, H, Miura, K, Abe, T, Koda, M, Yamazaki, M, Sankai, Y and Suzuki, K (2019) Muscle synergies during repetitive stoop lifting with a bioelectrically-controlled lumbar support exoskeleton. Frontiers in Human Neuroscience 13, 142. https://doi.org/10.3389/fnhum.2019.00142CrossRefGoogle ScholarPubMed
Theurel, J and Desbrosses, K (2019) Occupational exoskeletons: Overview of their benefits and limitations in preventing work-related musculoskeletal disorders. IISE Transactions on Occupational Ergonomics and HF 7, 264280. https://doi.org/10.1080/24725838.2019.1638331CrossRefGoogle Scholar
Theurel, J, Desbrosses, K, Roux, T and Savescu, A (2018) Physiological consequences of using an upper limb exoskeleton during manual handling tasks. Applied Ergonomics 67, 211217. https://doi.org/10.1016/j.ergon.2014.03.008CrossRefGoogle ScholarPubMed
Torricelli, D, Rodriguez-Guerrero, C, Veneman, JF, Crea, S, Briem, K, Lenggenhager, B and Beckerle, P (2020) Benchmarking wearable robots: Challenges and recommendations from functional, user experience, and methodological perspectives. Frontiers in Robotics and AI 7, 168. https://doi.org/10.3389/frobt.2020.561774CrossRefGoogle ScholarPubMed
Toxiri, S, Koopman, AS, Lazzaroni, M, Ortiz, J, Power, V, de Looze, MP, O’Sullivan, L and Caldwell, DG (2018) Rationale, implementation and evaluation of assistive strategies for an active back-support exoskeleton. Frontiers Robotics AI 5, 53. https://doi.org/10.3389/frobt.2018.00053CrossRefGoogle ScholarPubMed
Toxiri, S, Näf, MB, Lazzaroni, M, Fernández, J, Sposito, M, Poliero, T, Monica, L, Anastasi, S, Caldwell, DG and Ortiz, J (2019) Back-support exoskeletons for occupational use: An overview of technological advances and trends. IISE Transactions on Occupational Ergonomics and Human Factors 7(3–4), 237249. https://doi.org/10.1080/24725838.2019.1626303CrossRefGoogle Scholar
Van Engelhoven, L, Poon, N, Kazerooni, H, Barr, A, Rempel, D and Harris-Adamson, C (2018) Evaluation of an adjustable support shoulder exoskeleton on static and dynamic overhead tasks. Proceedings of the Human Factors and Ergonomics Society Annual Meeting 62, 804808. https://doi.org/10.1177/1541931218621184CrossRefGoogle Scholar
Voilqué, A, Masood, J, Fauroux, JC, Sabourin, L and Guezet, O (2019) Industrial exoskeleton technology: Classification, structural analysis, and structural complexity indicator. In 2019 Wearable Robotics Association Conference, WearRAcon 2019, pp. 13–20. http://dx.doi.org/10.1109/WEARRACON.2019.8719395CrossRefGoogle Scholar
von Glinski, A, Yilmaz, E, Mrotzek, S, Marek, E, Jettkant, B, Brinkemper, A, Fisahn, C, Schildhauer, TA and Geßmann, J (2019) Effectiveness of an on-body lifting aid (HAL® for care support) to reduce lower back muscle activity during repetitive lifting tasks. Journal of Clinical Neuroscience 63, 249255. http://dx.doi.org/10.1016/j.jocn.2019.01.038CrossRefGoogle ScholarPubMed
Wang, H-M, Le, DKL and Lin, W-C (2021) Evaluation of a passive upper-limb exoskeleton applied to assist farming activities in fruit orchards. Applied Sciences 11(2), 757. https://doi.org/10.3390/app11020757CrossRefGoogle Scholar
Wei, W, Shijia, Z, Yuxuan, X, Jihua, G and Xichuan, L (2020a) A hip active assisted exoskeleton that assists the semi-squat lifting. Applied Sciences 10, 2424. https://doi.org/10.3390/app10072424CrossRefGoogle Scholar
Wei, W, Wang, W, Qu, Z, Gu, J, Lin, X and Yue, C (2020b) The effects of a passive exoskeleton on muscle activity and metabolic cost of energy. Advanced Robotics 34(1), 1927. https://doi.org/10.1080/01691864.2019.1707708CrossRefGoogle Scholar
Zhou, YM, Hohimer, C, Proietti, T, O’Neill, CT and Walsh, CJ (2021) Kinematics-based control of an inflatable soft wearable robot for assisting the shoulder of industrial workers. IEEE Robotics and Automation Letters 6(2), 18. https://doi.org/10.1109/LRA.2021.3061365CrossRefGoogle Scholar
Figure 0

Figure 1. Occupational exoskeletons classification based on kinematic structure (a) and type of actuation (b). Concerning the kinematics, rigid-structure devices can be classified in anthropomorphic and nonanthropomorphic devices, whereas soft exosuits do not present any kinematic structure. Actuation types include passive, semi-active, and active systems. In the picture actuation types are depicted for an anthropomorphic device. Sketches are provided for upper-limb devices.

Figure 1

Table 1. Classification of occupational exoskeletons (OEs) according to kinematics, structure, and actuation; within each category, a nonexhaustive list of commercial and prototypical OEs is given

Figure 2

Figure 2. Frequency analysis of the objective and subjective metrics between laboratory and field studies for back-support and upper-limb exoskeletons. Acronyms: Technology Acceptance Model (TAM), NASA task load index (NASA TLX).

Figure 3

Figure 3. Sample size analysis of the studies using back-support and upper-limb exoskeletons.

Figure 4

Figure 4. Number of papers that were published in the field of exoskeleton evaluations in laboratory and field conditions.

Figure 5

Table 2. List of stakeholders on the demand side involved in the process of adoption of occupational exoskeletons (OEs)

Figure 6

Figure 5. Roadmap toward large-scale adoption of OEs.

Supplementary material: File

Crea et al. supplementary material

Crea et al. supplementary material

Download Crea et al. supplementary material(File)
File 95.9 KB