Frobt 09 813907

TYPE Review
PUBLISHED 03 January 2023

DOI 10.3389/frobt.2022.813907
Ergonomic human-robot
OPEN ACCESS collaboration in industry: A
EDITED BY
Erin K. Chiou,
Arizona State University, United States
review
REVIEWED BY
Claudio Castellini, Marta Lorenzini 1*†, Marta Lagomarsino 1,2†, Luca Fortini 1,2,
University of Erlangen Nuremberg,
Germany
Soheil Gholami 1,2 and Arash Ajoudani 1
Arto Reiman, 1
Human-Robot Interfaces and Physical Interaction Laboratory, Italian Institute of Technology, Genoa,
University of Oulu, Finland Italy, 2Neuroengineering and Medical Robotics Laboratory, Department of Electronics, Information
*CORRESPONDENCE and Bioengineering, Polytechnic University of Milan, Milan, Italy
Marta Lorenzini,
marta.lorenzini@iit.it
†
These authors have contributed equally
to this work In the current industrial context, the importance of assessing and improving
SPECIALTY SECTION workers’ health conditions is widely recognised. Both physical and psycho-
This article was submitted to Human-
social factors contribute to jeopardising the underlying comfort and well-being,
Robot Interaction,
a section of the journal boosting the occurrence of diseases and injuries, and affecting their quality of
Frontiers in Robotics and AI life. Human-robot interaction and collaboration frameworks stand out among
RECEIVED 12 November 2021 the possible solutions to prevent and mitigate workplace risk factors. The
ACCEPTED 26 August 2022
increasingly advanced control strategies and planning schemes featured by
PUBLISHED 03 January 2023
collaborative robots have the potential to foster fruitful and efficient
CITATION
Lorenzini M, Lagomarsino M, Fortini L,
coordination during the execution of hybrid tasks, by meeting their human
Gholami S and Ajoudani A (2023), counterparts’ needs and limits. To this end, a thorough and comprehensive
Ergonomic human-robot collaboration evaluation of an individual’s ergonomics, i.e. direct effect of workload on the
in industry: A review.
Front. Robot. AI 9:813907. human psycho-physical state, must be taken into account. In this review article,
doi: 10.3389/frobt.2022.813907 we provide an overview of the existing ergonomics assessment tools as well as
COPYRIGHT the available monitoring technologies to drive and adapt a collaborative robot’s
© 2023 Lorenzini, Lagomarsino, Fortini, behaviour. Preliminary attempts of ergonomic human-robot collaboration
Gholami and Ajoudani. This is an open-
access article distributed under the frameworks are presented next, discussing state-of-the-art limitations and
terms of the Creative Commons challenges. Future trends and promising themes are finally highlighted,
Attribution License (CC BY). The use,
aiming to promote safety, health, and equality in worldwide workplaces.
distribution or reproduction in other
forums is permitted, provided the
original author(s) and the copyright KEYWORDS
owner(s) are credited and that the
original publication in this journal is human-robot collaboration, ergonomics, human factors, human-robot interaction,
cited, in accordance with accepted collaborative robots, industry
academic practice. No use, distribution
or reproduction is permitted which does
not comply with these terms. 1 Introduction
It is common knowledge nowadays that workers, worldwide, are exposed to
occupational risk factors that may have negative effects on their physical and mental
health. Activities such as heavy material handling, repetitive movements and prolonged
awkward sitting impose physical burden on workers’ bodies, resulting in the so-called
musculoskeletal disorders (MSDs) (Pascual and Naqvi, 2008). Despite the extensive
prevention efforts of the industrial world and the regulatory bureaus, these remain the
most widespread work-related health problem in the European Union (EU). According to
the European Agency for Safety and Health at Work (EU-OSHA), approximately three
out of five workers suffer from an MSD, among which backache and upper limb pain are
Frontiers in Robotics and AI 01 frontiersin.org

Lorenzini et al. 10.3389/frobt.2022.813907
FIGURE 1
A conceptual illustration of a worker accomplishing an industrial task without (A) and with (C) a collaborative robotic solution. The structure of
this article is also described in the schema (B), highlighting (blue) the research themes that we focus on.
the most common (de Kok et al., 2019). Recent studies have also serve as a valuable tool to train the workforce. To enhance the
highlighted the importance of considering stress and ergonomic condition and awareness of the workers, researchers
psychosocial factors along with the aforementioned physical investigated several possible solutions that may be grouped in the
solicitations proposing a holistic approach. Indeed, results following macro-areas: 1) effective design of comfortable and
from the sixth European Working Condition Survey (EWCS) adjustable workstations (Shikdar and Hadhrami, 2007; del Rio
claim that 25% of European workers reported that their Vilas et al., 2013; Peruzzini et al., 2019; Bongiovanni et al., 2022),
occupations have negative impacts on their mental and 2) development of intuitive feedback interfaces that warn about
emotional state (Kubicek et al., 2019). Overall, besides the risks and hazards (Villani et al., 2018a; Kim et al., 2018b, 2021a),
harmful effects on workers themselves, physical and mental and 3) creation of advanced human-robot shared workstations
health problems may lead to impressive costs to enterprises for fruitful and ergonomic collaboration in hybrid environments
and society, being one of the most common causes of (Kim et al., 2018a; El Makrini et al., 2019). As conceptually
disability, sick leave and early retirement (Hassard et al., 2014; illustrated in Figure 1, the latter is the key topic of this article.
James et al., 2018). This last scenario falls within the general term human-robot
In light of this, it becomes crucial for enterprises, trade interaction (HRI) and is probably on the cutting edge among the
unions and regulating authorities to address in an objective industrial research topics. This article is focused on human-robot
manner the hazardous factors that may lead to physical and collaboration (HRC) as opposed to HRI since these two terms
mental distress among the workforce. Ergonomics studies and hold different meanings. Interaction determines any kind of
interventions can be classified into “microergonomics” and action that involves another human being or robot, who does
“macroergonomics.” The field of “macroergonomics” not necessarily profit from it. On the other hand, humans and
concentrates on designing overall work systems and robots collaborating on a shared task form a team. A team is
determining how effective the technological and personnel defined as a small number of partners with complementary skills
sub-systems are with respect to external demands (Hendrick who are committed to a common purpose, performance goal,
and Kleiner, 2002). In this review, we focus on the individual and approach. The same holds for human-robot teams where the
dimension of the workers, namely the analysis of worker partners are humans and robots, committed to reach a joint
postures, workplace productivity, work physiology and objective through collaboration. The advent of collaborative
biomechanics within the scope of “microergonomics.” It is robots (CoBots) broadened the application possibilities
worth specifying that logistical and organisational aspects, attracting the attention of research community. CoBots can
such as resources allocation, shift/turn planning, and outline indeed support their human counterparts in performing
of the working environment, are not deeply examined in the text, physical (e.g., relieving the workers from part of the effort
but their impact on “microergonomics” is presented in the while handling heavy loads (Brosque et al., 2020)), cognitive
discussion. (e.g., visualising alternative behaviours to reduce operators’
A thorough ergonomics assessment builds the foundation for mental stress (Krupke et al., 2018)) and hazardous (e.g.,
a safer, healthier and less injury-prone workplace, resulting in an handling chemical material (Liu and Wang, 2020)) operations.
overall improvement of operators’ well-being. The key objective This relationship positively impacts productivity, flexibility, and
is to identify the risk factors and to quantify them, which can the creation of new jobs instead of replacing workers. The

previous referred studies demonstrated a link between CoBots developments. The ergonomics research trends are focused in
and improved work conditions; however, the inclusion of Gualtieri et al. (2021b); however, the corresponding ergonomics
ergonomic criteria in the development and implementation of assessment and monitoring tools, and their effects on the HRC
these technologies is far from being well-known. In this article, setups are not highlighted in this survey. Moreover, an
the authors are interested in understanding studies that included interesting survey is presented in Berg and Lu (2020) that
ergonomics as a requirement in HRC systems, giving particular introduces the available user-interfaces for HRC but without
attention to the segmentation between physical and cognitive studying the inherent ergonomics factors. In summary, What is
ergonomics, and presenting an offline or online application. missing is a review paper that tackle the inclusion and integration
Multiple review articles on the integration of human factors of human ergonomics principles specifically into human-robot
and ergonomics in engineering and manufacturing processes collaborative solutions.
design (Kolus et al., 2018; Sun et al., 2019) and, more in general, In an attempt to fill in the gap in the previously mentioned
within Industry 4.0 (Kadir et al., 2019; Sgarbossa et al., 2020) have surveys, in this article, we mainly focus on the works that
been proposed. Critical socio-technical factors for the successful explicitly address human factors and ergonomics within HRC
implementation of Industry 4.0 were examined by Sony and Naik solutions. Among all the HRC possibilities, we specifically
(2019, 2020). On the other hand, multiple review articles exist on consider those frameworks in which a CoBot interacts with
HRC frameworks in industrial environments but focus primarily the human to accomplish a shared task, by adapting online1
on technology development and determining and minimizing its its behaviour to address the counterpart’s demands. As such,
intrinsic safety risks. An exhaustive review on HRC in industrial teleoperation systems and exoskeletons are not covered here. The
environment is provided by Villani et al. (2018a), with specific foremost objective of this article is to provide a narrative review
focus on issues related to physical (safety) and cognitive of the current state-of-the-art in HRC to improve online human
(intuitiveness-of-use) interaction. In fact, first, the safety ergonomics in the industrial sector, and to highlight the most
standards are recollected to discuss the permitted interaction important and promising research themes identified for both
level between human and the robotic agents based on the physical and cognitive ergonomics.
introduced measures. Second, they inspect the user-interfaces, The process we implemented to carry out the review is the
in the sense of cognitive workloads, claiming the traditional lead- following. We conducted an automatic search for papers that
through and offline programming are still the most used contain the selected keywords (which will be specified for each
interfaces in industrial practice, despite the rise of more subtopic/Section) in Google Scholar and Scopus, since they are
intuitive methods such as multi-modal interaction and the most well-known and used Database in the target audience.
extended reality technologies (e.g., virtual and augmented Existing ergonomics assessment tools were investigated starting
realities). They conclude the review by listing the from the earlier reports in the field. Instead, the papers on
commercially-available solutions, and their applications in the ergonomics HRC were selected from 2011 until the moment
industrial setups to improve the efficiency of the conventional of paper preparation (August 2022). We then meticulously
systems. Still, the roles and effects of HRC setups on the examined the list of potentially relevant papers and excluded
improvement of ergonomics, especially physical ergonomics, is those that did not explicitly use/study the topic of interest or
not discussed in this article. Another interesting survey is those that only mentioned it in the literature review part of the
presented in Kumar et al. (2020). The authors abstractly introduction. When multiple papers presented the same/similar
categorise the HRC setups into three main aspects, 1) idea, we selected the one published first, or in the case of evolved
awareness: level of perception using sensor information idea, we chose the journal version. The exceptions were some
coming from human-operator, robots, and workspace, 2) preliminary conference publications of the work that was later
intelligence: development of algorithms to achieve the robot’s evolved and published in a journal for the purpose of historic
actions and behaviours, and 3) compliance: dealing with the narrative. In addition, we used several more general ergonomics
management of human expectation and communication between and HRC papers to establish a context, introduce basic concepts,
the agents. In particular, safety, trust-in-automation, and and support our statements.
productivity factors are comprehensively discussed in the The rest of the paper is organised as follows. Section 2
introduced intelligence category covering the latest research provides an overview of the of the existing ergonomics
done in these regards. Nevertheless, here, the ergonomics assessment tools to evaluate both physical and cognitive
aspects are not investigated. The available mechanisms to workload. The available technologies to monitor the human
assure safety of CoBot systems in manufacturing is also
discussed in Bi et al. (2021); Zacharaki et al. (2020). In
addition, Castro et al. (2021) mostly focus on the current
HRC research trends and their future directions. They claim 1 With the term “online” we refer to methods and procedures that
respond in a timely manner, but with no particular limit on time
that better interactions, cognitive integration, and the presence of latency (which are instead required for “real-time” systems and can
effective metrics are the fundamental necessities of the future be ensured only following specific technological requirements).

FIGURE 2
Overview of trade-offs among the different categories of methods for the assessment of physical and cognitive workload. The variation in
thickness indicates the feature trend, while the colours reflect the positive (blue) or negative (red) impact.
state are presented in Section 3, which may be beneficial to • subjective judgements: self-questionnaires from workers or
automatise ergonomics evaluation. In Section 4, preliminary narrative interviews from experts;
attempts of ergonomic HRC frameworks are illustrated and in • systematic observations: collected on-site at the workplace
Section 5, state-of-the-art limitations and challenges are or from video recordings;
discussed. Last, the future trends are highlighted and the • direct measurements: performed on-site at the workplace
conclusions of the article are drawn in Section 6. or during simulations in laboratories.
Subjective measurements are surveys that can be completed

2 Ergonomics assessment in either by the tested subjects or by an interviewer but always
industrial settings reflect the point of view of the former. With systematic
observations, we refer to those procedures (e.g., worksheet to
Due to the alarming statistics on workers’ health conditions be filled in, parameters to be collected) that are carried out by
in the industrial sector, various methods and approaches were experts and are based on simple observations of the examined
developed, in the last decades, to assess the exposure to risks in subjects. In recent times, many of the methods belonging to this
the workplace. Chemical and environmental agents are not category have been automatised by leveraging the benefits of
covered here, but we tackle those short or long-term factors that sensor technologies, but their use is not strictly necessary.
induce a hazardous workload on the operators. This section Conversely, the techniques pertaining to direct measurements
provides an overview of the studies wherein tools to evaluate inevitably imply a sensor system due to the required accuracy and
ergonomics and human factors were developed and presented, online availability of the measurements. The ergonomics
categorising them according to the class of risk factors they seek assessment can then be extracted directly from the collected
to address. Specifically, the approaches that tackle physical data or estimated by integrating them within ad-hoc models. The
workload are illustrated in Section 2.1 while studies on the same categorisation may be applied and are adopted here for
psychosocial/organisational determinants, defined as cognitive load measurements. The trend of some peculiar
“cognitive” workload, are outlined in Section 2.2. To focus features, varying among these three groups in an orderly way,
the literature analysis on the assessment of workers’ ergonomics is represented in Figure 2.
in industrial settings, we combined terms associated with
workload (e.g., “physical,” “cognitive,” “workload,” “stress,”
“effort,” “workplace”) with terms related to its analysis (e.g., 2.1 Physical workload
“human factors,” “ergonomics,” “assessment,” “evaluation”)
and we discard the studies dedicated to office work or In the earliest studies on the assessment of physical
services sectors. Indeed, most of the literature concentrates workload (Winkel and Mathiassen, 1994; Westgaard and
on factory jobs. Earlier reports (Burdorf, 1992; van der Beek and Winkel, 1996), the authors introduced the term
Frings-Dresen, 1998) on the assessment of physical workload “mechanical exposure” to denote all the factors connected
conventionally categorised the methods in the following three to the biomechanical forces generated into the human body
groups: when performing a work task. In this review, we embrace the

FIGURE 3
Overview of the ergonomic assessment methods to address both physical and cognitive workload, applicable to industrial settings.
same concept, thus not considering the full physical working questionnaires or interviews (Kuorinka et al., 1987; Bigos
environment (lighting, noise, thermal environment, etc.). et al., 1991; Dickinson et al., 1992; Wiktorin et al., 1993),
Adopting the general model proposed by van der Beek and among which the National Aeronautics and Space
Frings-Dresen (1998), which describes how the working Administration—task load index (NASA–TLX) (Hart and
situation induces responses and health effects in the Staveland, 1988; Hart, 2006) is the most used. Almost all the
workers, we can distinguish two types of physical exposure: strategies for subjective judgements developed up to now were
external and internal exposure. External exposure refers to the built on the basis of the above-mentioned earliest attempts.
work environment and the actual working method, i.e. Traditionally, data were collected using written means, but
adopted postures, executed movements, and exerted forces more recent innovations include web-based facilities. The
that workers exploit to perform an activity with their methods based on subjective judgements have the benefit of
anthropometric characteristics. The corresponding being straightforward to use (no specific expertise is required),
moments and forces within the human body are instead the applicable to a broad range of work situations and appropriate
internal exposure. for surveying high numbers of subjects at comparatively low
Among the groups defined above and illustrated in cost. Nevertheless, they are vulnerable to many influences and
Figure 3, subjective judgements and systematic observations several studies have shown that they have too low validity
are determined to tackle external exposure. The vast majority (Burdorf and Laan, 1991) and reliability (Wiktorin
of tools employed in the current industrial scenario to assess et al., 1993) with respect to the demands for ergonomic
workers’ ergonomics relies on such two categories, which are interventions.
covered in Sections 2.1.1, 2.1.2. The methods that they gather
are presented building on some comprehensive reviews (Li 2.1.2 Systematic observations
and Buckle, 1999; David, 2005; Marras, 2006; Andreas and Several approaches were developed in the last decade to
Johanssons, 2018), in which ergonomics tools were listed, systematically record workplace exposure to be examined by
classified, and compared. On the other hand, direct an observer and stored on ad-hoc sheets. These are commonly
measurements can be employed to estimate internal referred to as “pen-and-paper”2 methods. Most of them have
exposure. The corresponding category is addressed in been conceived on the basis of the two most relevant and
Section 2.1.3. widespread normative aiming to establish ergonomic
recommendations for workers, i.e., the International
2.1.1 Subjective judgements Organization for Standardization (ISO) 11228 and the
To assess physical workload, body discomfort, or job stress, European Standards (EN) 1005. The postures and movements
it is possible to directly query the workers, investigating both
physical and psychosocial factors. These methods take the form
of body map (Corlett and Bishop, 1976), rating scales (Shackel
2 These methods are called “pen-and-paper” since they were originally
et al., 1969; Borg et al., 1985), checklists (Drury and Coury, conceived as hand-written means, but, at present, their digital version
1982; Cox and Mackay, 1985; Keyserling et al., 1992), and is available.

of the workers can be carefully evaluated by a number of developed to provide a unique and comprehensive ergonomic
indicators: Posturegram (Priel, 1974), Posture targeting analysis. The EAWS comprises different sections (postures and
(Corlett et al., 1979), ovako working posture analysing system movements, action forces, manual material handling, and upper
(OWAS) (Karhu et al., 1977), rapid upper limb assessment limb, respectively) whose outcome can be integrated into a final
(RULA) (McAtamney and Corlett, 1993), hand-arm- score. Similarly, with the composite ergonomics risk assessment
movement analysis (HAMA) (Christmansson, 1994), a (CERA) (Szabò and Dobò, 2018) technique, a unified evaluation
method assigned for the identification of ergonomics hazards can be obtained after a separate determination of the different
(PLIBEL) (Kemmlert, 1995), quick exposure check (QEC) (Li ergonomic risks, also based on workplace history. Lastly, the key
and Buckle, 1998), and rapid entire body assessment (REBA) indicator method (KIM) was introduced to tackle manual
(Hignett and McAtamney, 2000). These techniques are relatively handling operations (KIM-MHO) (Klussmann et al., 2017),
inexpensive to carry out and can be used in different work lifting, holding and carrying (KIM-LHC) and pulling and
situations without hindering the workers, but they are pushing (KIM-PP) (Steinberg, 2012).
applicable only to rather static or repetitive jobs. Nevertheless, all these pen-and-paper techniques must be
Besides postures, other workplace factors such as load/force, employed by trained experts as an offline procedure after
repetition, duration of movement, vibration, and their collecting observations/recordings, which is rather time-
interaction/combination have to be considered. Similar to consuming and does not provide immediate results. For this
these techniques for posture analysis, muscle fatigue analysis reason, several attempts were made to automatise the completion
(MFA) (Rodgers, 2004) was proposed, whereby each body part is of some of the worksheets mentioned above to perform an online
scaled into four effort levels according to its working position, but ergonomics evaluation. Specifically, the RULA (Shaikh et al.,
also to the duration of the effort and the frequency. Considering 2003; Ray and Teizer, 2012; Vignais et al., 2013; Haggag et al.,
lifting and carrying loads, National Institute for Occupational 2013; Puthenveetil et al., 2015; Plantard et al., 2017), the REBA
Safety and Health (NIOSH) equations (Waters et al., 1993) were (Busch et al., 2017; Van de Perre et al., 2018) and the EAWS
introduced to define the suggested load weight limit to be lifted (Bortolini et al., 2018; Malaise et al., 2019) were considered,
by human operators considering gender, forces exerted on the respectively, and combined with human motion data in both
spine structure, and calories consumed during the effort. The virtual and real environments. However, the main limitation of
Washington Industrial Safety and Health Act (WISHA) lifting the observational methods still stand, i.e., the dynamics of the
calculator was then developed based on NIOSH. Mechanical tasks are considered to a limited extent (e.g., interaction forces
exposure can then be evaluated with respect to intensity (or are considered constant).
magnitude), repetitiveness and duration even with the strain
index (SI) (Steven Moore and Garg, 1995) and its revised versions 2.1.3 Direct measurements
(Garg et al., 2017), the American Conference of Governmental To address humans’ physical internal exposure, direct
Industrial Hygienists—threshold limit values (ACGIH–TLV) measurements collected on the human subjects through
(ACGIH, 1981), an assessment technique for postural loading suitable sensor systems were generally integrated with more or
on the upper body based on joint motion discomfort and fewer complex models of the human body. Several algorithms
maximum holding time (LUBA) (Kee and Karwowski, 2001), were proposed for estimating muscle tensions and joint loads
hand arm risk assessment method (HARM) (Douwes and using detailed models of the human musculoskeletal system. One
Kraker, 2009) and manual handling assessment charts (MAC) of the most well-known is the algorithm underlying the open-
(Monnington et al., 2003). Snook and Ciriello (1991) proposed a source software “OpenSim” (Delp et al., 2007). This platform
detailed procedure to assess the exerted force to perform pushing enables the creation of dynamic simulations of movement that
and pulling activities, taking into account the weight/distance of integrates off-the-shelf models describing the anatomy and
the handled object, the frequency, and duration of the action. physiology of the elements of the neuromusculoskeletal
Finally, some methods were focused on actions performed at system and the mechanics of multi-joint movement. Similar
high frequency with low loads and consider even the recovery capabilities are offered by the simulation software “Anybody”
time like occupational repetitive actions (OCRA), a concise index (Damsgaard et al., 2006), which is capable of analysing the
for the assessment of exposure to repetitive movements of the musculoskeletal architecture of humans as rigid-body systems.
upper limb (Occhipinti, 1998). Although these indices are more Hence, standard methods of multi-body dynamics (i.e., inverse
exhaustive and have been widely adopted both by practitioners kinematics and inverse dynamics) can be applied but integrating
and researchers, they lack of precision and reliability and into the model a reasonable representation of the muscle
subjective variability can influence their results. Moreover, geometry and the recruitment pattern of the muscles. An
they do not provide a consistent and overall measure of the analogous package is virtual interactive musculoskeletal
ergonomic risk since every index focuses on a specific manual system (VIMS) (Chao et al., 2007). Besides the massive
material activity. In view of this, the ergonomic assessment studies behind the development of these platforms, there are
worksheet (EAWS) method (Schaub et al., 2013) was also some minor works whereby muscle models were introduced

to account for internal exposure (Bhargava et al., 2004; Forster, by traditional ergonomic tools, while overcoming the limitation
2004; Nakamura et al., 2005; Fraysse et al., 2009; Millard et al., of the laboratory-based methods, impractical and hardly
2013). All the above-mentioned platforms and methodologies customisable. For instance, Maurice et al. (2017) proposed
account for joint reactions (forces and torques) from motion data multiple ergonomic indicators that are capable of quantifying
by using inverse dynamics and then optimisation techniques to exhaustively and concisely the physical demands endured by a
compute the muscle tensions. Nevertheless, due to the actuation worker when executing various manual activities, addressing
redundancy (the number of muscles is greater than the number both postures/movements and forces/torques. By exploiting
of degrees of freedom (DoFs) of the system), a desired motion in the principles of humanoid robotics to model human
terms of joint torques can be achieved by an infinite number of kinodynamics, Lorenzini et al. (2022) proposed an online
activation patterns of the muscles. Another drawback is that the multi-index approach to account for multiple potential
complex musculoskeletal models underlying require the contributors to MSDs, also giving importance to the subject-
identification of numerous parameters (Ayusawa et al., 2014). specific requirements of the workers. In the same line, Gholami
Alternatively, they can be obtained by means of anthropometric et al. (2022) introduced a set of quantitative metrics to take into
standards and tables (Herman, 2007; Winter, 2009) thus they are account operators’ ergonomics in the assessment of teleoperation
not subject-specific. interfaces. Still chasing a reduced-complexity approach but
An alternative solution is to measure muscle activation using adding a certain level of accuracy for more sophisticated body
electromyography (EMG) directly and exploit empirical models districts, Ventura et al. (2021) developed a flexible model of the
(Hill, 1938; Stroeve, 1999) to convert such activation into muscle human spine mechanics for assessing compressive loading.
tensions (Buchanan et al., 2004). One of the first attempts at Finally, Latella et al. (2019) presented a stochastic
using an EMG-based technique for the monitoring of low back methodology for the simultaneous floating-base estimation of
physical exposure and cumulative compression was made by the human whole-body kinematics and dynamics toward online
Mientjes et al. (1999). Afterwards, experimentally recorded EMG ergonomics assessment. Nevertheless, this approach is thus far
signals have been used in several studies to directly drive rather limited.
simulations of upper (Manal et al., 2002; Village et al., 2005;
Nikooyan et al., 2012; Pau et al., 2012) and lower (Lloyd and
Besier, 2003; Kumar et al., 2012; Manal et al., 2012; Sartori et al., 2.2 Cognitive workload
2012; Meyer et al., 2017) extremity musculoskeletal models.
Whole-body muscle tension estimation by means of optical The evidence that excessive cognitive demand at work can
motion-capture and EMG measurements was enriched with a harm the workers’ health and performance has led to a renewed
visual feedback interface in (Murai et al., 2010). Nevertheless, interest in cognitive load theory (CLT). CLT examines the
even EMG-based approaches incorporate numerous parameters interaction of cognitive structures, information and its
and the use of EMG presents several drawbacks. The correct implications (Sweller et al., 1998). Precisely, the term
placement of EMG sensors is quite difficult and their relative “cognitive load” refers to the amount of processing that
movement in dynamic conditions makes the estimates performing a particular task imposes on the learner’s
questionable (Farina and Merletti, 2001). EMG signals are cognitive system (Paas et al., 2003). Xie and Salvendy (2000)
inevitably affected by various noise signals or artifacts (De provide a detailed conceptual framework of human information
Luca et al., 2010). Finally, many EMG-based techniques are processing. Definitions of instantaneous load, peak load,
conceived for specific body parts. As such, due to the inner accumulated load, average load, and overall load are presented
complexity and reduced practicability, the methods based on to investigate the trend of cognitive load over time as a response
direct measurements have been implemented nearly entire in to stimuli that an activity and/or environmental conditions are
laboratory settings. imposing on the subject. The following sections investigate the
Aiming to meet contemporary industry demands, some current state-of-the-art about mental workload modelling and
innovative approaches have been recently proposed. The latter cognitive cost estimation for performing tasks, according to the
evaluate human physical workload by relying on the online three main categories defined at the beginning of Section 2 and
monitoring of human kinodynamic3 state through reduced- displayed in Figure 3.
complexity estimation algorithms. The objective is to also
account for the workers’ internal exposure, which is neglected 2.2.1 Subjective judgements
Thus far, narrative interviews and subjective rating scales
represent the most commonly used method to measure the
cognitive load both in laboratory and industrial settings
3 The term “kinodynamic” measurements is used, in this article, to (Rubio et al., 2004; Leppink et al., 2013). NASA-TLX (Hart
indicate the variables associated with both human kinematics
(i.e., positions, velocities, and accelerations) and dynamics and Staveland, 1988; Hart, 2006), subjective workload
(i.e., quantities related to interaction forces). assessment technique (SWAT) (Reid and Nygren, 1988) and

subjective workload technique (SWORD) (Vidullch et al., 1991) people relative to their attention distribution, decision-making,
are just a few of the available questionnaires in the literature. Self- mental overload, frustration, stress and errors. Head pose
ratings nevertheless have many limitations (Naismith et al., estimation and skeleton tracking were exploited to
2015). The main drawback is the assumption that people are investigate the workers’ attention and assess hyperactivity
able to introspect on the cognitive processes and report the and unforeseen movements. Despite the growing interest in
amount of experienced cognitive effort (Naismith et al., 2015). the topic, assessing cognitive load through visual monitoring
Additionally, they are often affected by many biases, such as systems is a moderately new topic (Bisogni et al., 2022) with the
acquiescence and social desirability, and their outcome may potential to bring solutions from the laboratory to the actual
appear questionable. Lastly, they work offline and the deep shop floor.
comprehension of collected data requires specific skills in the
field of cognitive ergonomics and cognitive science. 2.2.3 Direct measurements
More recently, researchers presented tools intended to be A great deal of previous research into cognitive load
used directly by workers involved in the manufacturing domain. assessment explored direct measurements of physiological
For instance, the work in Thorvald et al. (2019) presented a factor signals. Physiological measurement of workload relies on the
assessment tool and a handbook, denoted cognitive load physical reaction of the human body to an intense mental
assessment for manufacturing (CLAM), to estimate the mental demand (Sweller et al., 1998). The monitoring of brain
workload that human operators were expected to employ within activity is the most direct and accepted form to investigate
specific assembly tasks and workstation layouts. End-users were cognitive processes. The electroencephalogram (EEG) provides
asked to reflect on different aspects of their daily activity and rate an online, continuous measure of fine fluctuations in
factors on a scale from 0 to 8. A specific combination of them instantaneous mental load (Al-Shargie et al., 2016; So et al.,
resulted in the final cognitive load score. 2017; Yang et al., 2019). Nevertheless, motion artefacts and noise
due to electrical interference, breathing and heartbeat make the
2.2.2 Systematic observations EEG signal not deployable in industrial settings.
To assess the cognitive effort in performing industrial Work-related stress and strain were found to alter
activities, two strategies have been adopted based on direct sympathetic-parasympathetic nervous system balance and
observations of the involved operators: 1) performance raise the risk of heart diseases (Hughes et al., 2019).
measures on either the first or secondary task and the 2) Therefore, the short-term cardiovascular consequences of
analysis of behavioural characteristics and modifications. mental work were investigated. In the literature, several heart
Task- and performance-based techniques involve rate variability (HRV)-derived metrics were defined in the time,
measurements on both the primary and secondary tasks. The frequency and non-linear domain. Typically, cognitive workload
idea behind is that people have limited resources, so tasks leads to a decrease of time-domain measures (e.g. mean RR
performed concurrently are supposed to reflect the level of the intervals (Henelius et al., 2009)), as well as a reduction of low
cognitive load imposed by the primary task (Kaber and Riley, frequency (LF, 0.04 − −0.15 Hz) and high frequency (HF,
1999; Paas et al., 2003; Wu and Li, 2013). State-of-the-art 0.15 − −0.4 Hz) powers, while the ratio LF/HF increases
measurements to assess the performance are task time, run (Durantin et al., 2014; Delliaux et al., 2019).
time, reaction time, accuracy, and error rate. Despite the high Besides, the galvanic skin response (GSR, also known as
sensitivity and reliability, this technique interferes considerably electrodermal activity, EDA) has been widely studied for
with the usual task execution, making it rarely applicable, even in indexing variations in sympathetic arousal associated with
laboratory settings. emotion, cognition, and attention (Critchley, 2002; Poh et al.,
In an effort to design less obstructive monitoring systems 2010; Setz et al., 2010). GSR or EDA is the measurement of the
and maximise users’ comfort, the applicability of external continuous changes in the skin’s electrical conductance arising
sensory systems has been recently examined. Early studies from the diverse sweating activity of the human body.
exploited camera sensors for automatic emotion recognition Researchers identified two components in high-resolution
(Glowinski et al., 2011; Karg et al., 2013; Kleinsmith and EDA signal, i.e., the tonic (skin conductance level, SCL) and
Bianchi-Berthouze, 2013; Roy et al., 2020) and expression phasic response (skin conductance response, SCR), and used
synthesis (Karg et al., 2013; Kleinsmith and Bianchi- derived metrics to quantify cognitive states and stressful periods
Berthouze, 2013), as well as activity-related behavioural (de Santos Sierra et al., 2011; Kyriakou et al., 2019; Han et al.,
indexes of stress (Giakoumis et al., 2012; Aigrain et al., 2020).
2015). A quantitative and online framework was developed More recent studies also included measures of respiratory
by Lagomarsino et al. (2022b) to monitor the cognitive activity (e.g., respiratory rate, volume and concentration of CO2
workload of human operators by detecting patterns in their in airflow) (Grassmann et al., 2016), eye activity (e.g., eye blink
motion directly from the input images of a low-cost RGB-D rate, intervals of closure, horizontal eye movement, pupil dilation
camera. The method examined how industrial work affects and eye fixation) (Ahlstrom and Friedman-Berg, 2006; Coyne

FIGURE 4
Overview of the different available technologies for the measurement of kinodynamic variables as well as physiological parameters to account
for human ergonomics.
and Sibley, 2016), cortisol level (Carrasco and de Kar, 2003), and kinodynamics data and (ii) biosignals and physiological
speech measures (e.g., pitch, rate, loudness, jitter and shimmer) indicators (see Figure 4). In each group, the choices of sensors
(Yin et al., 2008). and devices impose challenges that should comply with the rigid
Psychophysiological measurements provide objective and rules imposed in industrial settings, namely:
quantitative information and permit the visualisation of
continuous trends and the identification of detailed • to guarantee task timing and data synchronization
patterns of load. Nevertheless, signal acquisition requires requirements (online measurement);
expensive and impractical systems that are highly sensitive • to ensure workers’ safety and avoid excessive
to human movements and often hinder users’ daily activities. encumbrance, as well as physical and mental demand
For the above-mentioned reasons, the adoption of the (non-invasiveness4);
technology in real-world scenarios is subject to certain • to ensure the quality of measurement (accuracy and
limitations. precision).
It can be noted that these indicators have commonalities with

3 Human monitoring hardware and those introduced in Figure 2 about the workload assessment. The
systems challenges and the environmental limitations are in fact similar.
It is not our interest to address the economical cost comparison
In the previous sections, we went through the current of the devices market panorama.
standards and methods for ergonomics assessment applicable
to the industry. All the presented indexes and algorithms need, to
a certain extent, monitoring of human state. We then inspect the 3.1 Body kinodynamics
literature by searching for human monitoring hardware and
systems, by first selecting keywords related to broad concepts Biomechanical analysis requires data on the posture and forces
(e.g., “human monitoring,” “human sensors,” “biosensors,” exerted by a specific worker during their duties. It means that online
“biosignals”) and then specific categories (e.g., “motion-
capture,” “electrocardiography,” “electromiography,”
“electroencephalography”). Within the scope of this review
article, we divide the monitoring systems into two separate
4 With the term “non-invasiveness” is intended the absence of physical
groups, based on the measured quantities, namely (i) body constraints or interference to the subject.

monitoring of body posture and interaction forces is required to despite their compelling accuracy performances (Menolotto
estimate the time series data of physical workload. et al., 2020). This major drawback can be overcome using the
Human postures are commonly obtained by motion-capture latest advancement in markerless depth-based optical MoCap
(MoCap) systems. Indeed, realistic skeletal simulation is required to ™
systems (e.g., RealSense , Inter Intel Corp.). Accordingly, the
perform the synthesis and analysis of the performed human motions users can freely perform their activities and tasks without wearing
in time. In the synthesis phase, MoCap data leads to improvements a suit or having attached markers on their bodies. Nonetheless,
in the human model rendering, while the analysis aspects help the up to now, they offer less accurate measurements with respect to
researchers in obtaining critical insights into human musculoskeletal the marker-based systems, and they similarly suffer from the
systems, such as body joints angles and velocities, movements of occlusion problem. Above all, optical systems in general induce
body center of mass, body segments poses, etc. Different visibility issues due to the limited range of installed cameras. As
technologies and solutions have been developed to capture an alternative, IMU-based systems can track a variety of postures
motion, hereafter we analyse some of the most widespread in the cluttered environment associated with both indoor and
examples in the literature along with some industrial outdoor applications, indicating higher portability and
applications. Camera-based systems with infrared (IR) cameras deployability. In fact, it is not necessary to place or install any
can be used to triangulate the location of retroreflective rigid fixed infrastructure to use the inertia-based solutions. However,
bodies (markers) attached to the targeted subject (Nagymáté and drift, i.e., divergence of the output values from their real values
Kiss, 2018; Chatzitofis et al., 2021; Hu et al., 2021). In addition, that happens in time, is the common issue with the IMU-based
systems based on inertial measurement units (IMU) that track the systems that does not ensure an accurate absolute position of the
relative movements of articulated structures have become popular limbs (Damgrave and Lutters, 2009). Besides, artifacts due to skin
for their versatility (Vignais et al., 2013; Caputo et al., 2018; Marín movement can act as sources of measurement noise during the
and Marín, 2021). Moreover, at the time of writing, markerless acquisition of both IMU-based and marker-based systems, which
optical MoCap systems undergo significant research progress with can lead to errors that, in some cases, are of the same order of
high application potential. These systems rely on image processing magnitude as the motions of the studied joints.
and deep learning techniques to track human skeletal information On the other hand, to measure the force exerted by a worker,
online using off-the-shelf and relatively cheap RGB-D cameras force sensors and force plates are mainly applied (Hoffman et al.,
(Bortolini et al., 2020; Kim et al., 2021c). 2007; Arjmand et al., 2009). Force sensors can measure the
Let us now list some literature examples of MoCap applications interactions between workers’ hands and the tasks’ objects. In
in industry. The aforementioned MoCap technologies were addition, most of the available force plates can simultaneously
exploited, for example in Maurice et al. (2019), where a dataset measure the external ground reaction forces (GRFs) in three
of human motions in industry-like activities, fully labelled in line planes, i.e. vertical, anterior-posterior, and medial-lateral.
with EAWS, is presented. First, this dataset is applicable for However, the installation of the former and the portability of
classifying, predicting, or evaluating human motions in the the latter are issues that undermine their industrial adoption. As
industrial environment. Second, it supports the robotics a solution, wearable alternatives are suggested in the form of
communities to provide collaborative solutions aiming at gloves (Park et al., 2019) and shoes (Bamberg et al., 2008;
improving the workers’ ergonomics. In this study, the Xsens Muzaffar and Elfadel, 2020) equipped with force/torque
MVN Link and Qualisys were used as inertial and marker-based sensors. Instrumented gloves, for example, remove the need to
optical MoCap systems. Besides, the data collection procedure was equip handles and tools with force sensors or pressure mats.
recorded with two video cameras to be further analysed with the However, force sensor mats embedded within gloves acquire only
OpenPose library (Cao et al., 2021). A similar logistics-dataset was normal forces, require calibration and may shift during
also presented in Niemann et al. (2020) where picking and packing measurements (Ranavolo et al., 2018). Notably, the wearable
scenarios were recreated to be used in recognition and analysis of insole pressure system can acquire the GRF and plantar pressure
human activities in logistics applications. data under the foot. Moreover, they can be easily inserted or
Currently, the industrial exploitation of these technologies is attached to workers’ safety boots with a minimal hindering level.
challenging due to their inherent limitations (Mündermann et al.,
2006; Damgrave and Lutters, 2009; Bailey and Bodenheimer,
2012; Lopez-Nava and Angelica, 2016; Patrizi et al., 2016; Colyer 3.2 Biosignals and physiological indicators
et al., 2018; Yahya et al., 2019; Menolotto et al., 2020; Kanko et al.,
2021). The need for highly specialised equipment, regular As we stated in Section 2.2, investigating physiological signals
calibration routines, limited capture volumes, inconvenient can be helpful in understanding physiological-related aspects of
markers or specialised suits, as well as the significant work, such as mental stress, mental fatigue, and physical stress.
installation and operation costs of these systems, has so far The increased interest in these topics aroused from the recent
greatly impeded the adoption of the optical marker-based technological progress, making available wireless, off-the-shelf,
™
systems (e.g., OptiTrack , NaturalPoint, Inc.) in industry lightweight, and affordable biosensors.

Biosensors can be classified as 1) traditional biosensors; 2) duration, fixation rate, the fixation to saccades ratio, average
portable biosensors; 3) noncontact biosensors. saccade distance and velocity, peak saccade velocity, number of
Traditional modes of monitoring physiological quantities long fixations, and average pupil diameter (Ahlstrom and
rely on hospital-based diagnostics devices that, in a controlled Friedman-Berg, 2006). Online processing of the
environment, guarantee high-quality signals. However, they aforementioned metrics does not require significant
usually require trained staff and in most cases an offline post- computational power and is not as complex as EEG or other
processing phase to extract meaningful features. In this context, brain imaging techniques.To recap, with the aim to assess
more portable wearable biosensors emerged with the specific workers ergonomics in industrial environments, the preferred
attempt to extend the biosignals’ gathering to nonclinical sensor systems should be lightweight, easy to wear and/or set up
settings. Recent advancements in technology made available a without hindering workers’ activities and ensuring user comfort
vast number of biosensors convenient (e.g., headsets, chest straps, even for prolonged usage. Concerning body kinodynamics,
and wristbands) that can be deployed to check the ergonomic inertial-based devices provide accurate measurements but may
state of the workers in the field. be impractical in some task conditions. On the other hand,
Chest-strap sensors were primarily used for monitoring the external sensor systems, such as cameras, offer non invasive
cardiac activity (e.g., HRV) of workers. Chest straps offer an easy analysis of the human motion but suffer from occlusion and
and accurate alternative to traditional electrocardiography feature limited accuracy. Besides, biosignals and physiological
(ECG) measurements. Although progress in technology design indicators provide useful insight about the human state. These
has reduced the bulkiness of the sensor on the strap, the have shown a good potential for offline validation of possible
acceptance and practicality of such a device are still in their ergonomic interventions, but may be not appropriate for online
initial stages (Hinde et al., 2021). As stated in Section 2.2, a few application.
researchers attempted to assess workers’ mental stress and
cognitive load based on brain waves collected from EEG
headsets. The main advantage of these wearable devices is that 4 Ergonomics in human-robot
they show compelling rapidity performances in pointing out collaboration
changes in workers’ mental state fulfilling the online
requirements. Nevertheless, capturing high-quality EEG This section presents the online compensatory measures and
signals in the field is more challenging compared with other strategies that a robotic partner, fed with the ergonomic
physiological signals due to several intrinsic artifacts (e.g., eye evaluation of an operator (Section 2) and/or with the data
blinking and facial muscle movement) (Mijović et al., 2017). collected with available monitoring technologies (Section 3),
Among different wearable biosensors, wristband-type biosensors can put in place to mitigate the human workload. To focus
allow researchers to acquire multiple physiological signals (e.g., the literature analysis on ergonomic collaborative robotics, we set
photoplethysmography, EDA, and skin temperature) without the conceptual boundaries on the terms describing HRC in
interrupting workers’ ongoing tasks. However, measuring relation to the industrial sector (e.g., “human-robot,”
physiological signals using a wristband-type biosensor at the “collaborative robot,” “manufacturing,” “automation”) and
industrial floor is still challenging because of the large number of human ergonomics (e.g., “physical,” “cognitive,” “workload,”
extrinsic signal artifacts and distortions that come from workers’ “stress,” “effort”). The polar plot in Figure 5 illustrates the
movements, sensor displacement, environmental noises, and the production of papers linked to our research query over a ten-
lower quality of sensor electrodes compared with wired year time window (from 2011 to 2022).
biosensors (Jebelli et al., 2019).
Finally, we have noncontact sensors that are able to acquire
psychophysical signals without any contact between the sensor 4.1 Physical ergonomics in human-robot
probe and the human body, guaranteeing the best performances collaboration
in terms of non-invasiveness. Pupil diameter, gaze data, gaze
duration, and eyelid closure patterns can be remotely recorded Within the scope of this review article, we framed the
through infrared eye-gaze tracking systems. Eye-tracking is a different approaches to address workers’ physical ergonomics
functional and highly neuro-ergonomic solution for gathering during HRC in two macro areas. On the one hand, several
both mental workload and relevant practical attentional authors elected some among the systematic observations
information. It is a non-invasive and easy to set up device, methods presented in Section 2.1.2 to either identify a more
allowing for consistent data collection. Moreover, it does not ergonomic human posture or define a cost for task planning and
impose any physical burden on users, and the calibration routine role allocation. Hence, the criteria to drive HRC were selected
is fast and straightforward. A variety of measures were described directly among the gold standard ergonomic tools. On the other
in the literature that elucidates the efficiency of visual search hand, assumptions were made by some researchers to indirectly
related to mental workload, including fixation count, fixation achieve more ergonomic conditions for the workers, e.g. the

FIGURE 5
Number of papers related to the main assessment variables in physical and cognitive ergonomics.
reduction of human joint torque/fatigue or the optimisation of the human was induced to assume his/her most natural and
the human field of view/arm kinematics would improve task ergonomic posture according to REBA. Finally, El Makrini et al.
ergonomics. For sake of readability, in the following, these two (2019) employed the REBA to set a criterion for task allocation in
categories are addressed as standard-based and cost-based a human-robot assembly operation. Each sub-task of the
approaches. assembly was assigned to either the robot or the human based
on the task requirements, the capability of the agents but also the
4.1.1 Standards-based approaches evaluation of the human body posture given by REBA.
The works that deem ergonomics in HRC in a direct way are Right after the REBA, the most commonly used standard
presented and grouped in the following according to the ergonomics tool within HRC is the RULA method, likewise
observational method employed. Above all, the REBA method posture-based. Ferraguti et al. (2020) employed RULA within
was exploited by a bunch of researchers. Busch et al. (2017) first the proposed architecture for human-robot co-manipulation.
adopted REBA to improve the workers’ comfort and safety Different objects were considered during the experiments and
during a human-robot collaborative task. The human posture positioned in the most comfortable way for the user to operate
that minimised the score returned by REBA was computed and them, estimated based on the postures with the lower RULA
the robot pose was adjusted online to let the human perform the score. In addition, the position of such objects could be
task in the optimised body configuration. A graphical interface conveniently adjusted directly by the user thanks to the robot
was also developed here to inform the user about the ergonomics admittance control. Similarly, Shafti et al. (2019) presented an
evaluation results. The differentiable version of REBA was then approach to continuously invoking cooperative robot
introduced in a later work (Busch et al., 2018) to simultaneously movements that meet the human partner’s ergonomic
design robot motion (in the same way as before) and plan the postures according to RULA. Conversely, a simulation
sequence of actions in the task. The REBA technique was also presented in Zacharias et al. (2011) adopted RULA not to
adopted in Van de Perre et al. (2018) to both predict and optimise identify ergonomic configurations for the humans but to plan
human ergonomics during an hybrid co-carrying task. First, the a more human-like robot motion, which appeared safer and
human postures from hands poses were computed for a set of more interpretable from the human point of view. Human-robot
states (i.e., possible configurations to manipulate the object) and role allocation is instead addressed by Merlo et al. (2021), which
an ergonomic cost according to REBA was assigned to each state. proposed a RULA-based model for physical risk prediction. The
Next, the sequence of states was found that minimise this cost, developed online strategy can assign actions among the agents
resulting in a joint plan for the two agents that was optimised for (i.e., human or robot) according to a human physical state
human ergonomics. On the other hand, Zanchettin et al. (2019) indicator, called kinematic wear, that can account for the
presented a control strategy to facilitate the human to assume a usage of each joint based on RULA guidelines. From the same
more convenient body configuration while operating on a bulky research group, Fusaro et al. (2021) proposed a method to
object held by the robot. The robot moved the workpiece so that generate robot plans for both autonomous and human-robot

cooperative tasks taking into account human ergonomics. The the adaptive robot behaviour at the onset of physical fatigue.With
developed approach allowed the robot to online adapt its plan to the aim to plan a shared human-robot assembly task, in Michalos
the human partner, by choosing the tasks that minimize some et al. (2018) a multi-criteria method was proposed that employs a
execution costs that entailed human availability, decisions, and set of quality, productivity but also ergonomics criteria to identify
ergonomics. The latter was specifically addressed by using the the best planning scenario in a simulated environment. A fatigue
WISHA index which, unlike REBA and RULA, considers also the model as well as the NIOSH index were exploited here as costs for
task frequency, duration and the involved weights. Task planning the search algorithm. Lamon et al. (2019) introduced a set of
for human-robot collaborative tasks was also addressed in Faber indexes to evaluate agents performance combined with an offline
et al. (2016), Liau and Ryu (2020) and Pearce et al. (2018), by allocation algorithm for optimal role allocation in an industrial
determining the ergonomic cost with the OWAS, the RULA and assembly task. Among these indexes, agent dexterity and effort
the SI, respectively, but in simulation or offline, at a later stage. were proposed to consider human comfort and physical fatigue
Lastly, it is worth mentioning some studies in which thus ergonomics from both a kinematic and dynamic point of
observational methods were employed to evaluate HRC view, respectively. In the last two studies, the fatigue model was
workstations offline (Gualtieri et al., 2020; Colim et al., 2021; based on the one proposed in Ma et al. (2009).
Dimitropoulos et al., 2021; Palomba et al., 2021) or in simulation The multiple ergonomic indicators defined by Maurice et al.
(Castro et al., 2018; Heydaryan et al., 2018; Lietaert et al., 2019; (2017) were employed to orient the design of a CoBot, but the
Mateus et al., 2019; Laudante et al., 2020). proposed set of variables, due to their ease of computation, may
be employed online for HRC. Similarly, in Rapetti et al. (2019) an
4.1.2 Cost-based approaches optimisation problem was formulated, where ergonomics targets
The studies that indirectly address human ergonomics such as muscular effort and body posture were mapped to human
demands in HRC are illustrated hereafter, grouped according kinodynamic quantities such as joint torque and joint angles/
to the quantity employed as a cost. In Bestick et al. (2016, 2015) velocities, to develop a human-aware robot controller for HRC.
the most convenient robot configurations when handing over an Alternatively, in Marin et al. (2018) the optimisation of
object to a human partner were investigated. A simple cost “contextual ergonomics models” was proposed to successfully
function was assumed a priori considering the distance of reduce the muscle activation of subjects performing a drilling
human joints from a neutral position. The higher this task. The presented models were Gaussian process latent variable
ergonomic cost, the less likely the associated human body models trained offline with detailed musculoskeletal simulations
configuration, leading to a specific robot pose. Instead, in but can be employed in a low-dimensional latent space, featuring
Bestick et al. (2018), the ergonomic cost was learned online potentially online capabilities.
via Bayesian inference, based on implicit physical queries from Lastly, human effort can be expressed through physiological
the robot. Katayama and Hasuura (2003) introduced five measures. In this view, Peternel et al. (2017, 2018) presented a
different optimisation models to characterised human method to adapt online the robot behaviour to human fatigue,
comfort. Among the latter, the “medium joint angle index” which was modelled based on human muscle activity measured
was selected by Sisbot and Alami (2012) to model arm with EMG sensors.
comfort, which, along with human safety and visibility, was
employed as a criterion to generate ergonomic robot paths
toward the object transfer point (OTP) for handover. The 4.2 Cognitive ergonomics in human-robot
extensions of this human-aware manipulation planner took collaboration
into account also HRC constraints (Mainprice et al., 2011) as
well as human mobility (Mainprice et al., 2012). In addition to ensuring the operator’s physical safety
In these studies, the ergonomic cost selected only consider and comfort, the cognitive resources demand and mental
the kinematics of the human actions. On the other hand, the stress induced by the close interaction with a CoBot should
“joint torque model” proposed in Katayama and Hasuura (2003) not be overlooked. Early interviews with the potential future
was instead employed by Parastegari et al. (2017) to predict an users (i.e., actual industrial workers) of these robots
OTP that matches the human receiver preferred position. The revealed controversial attitudes and social cues
overloading joint torque, i.e., the torque induced only by the (Wurhofer et al., 2015; Elprama et al., 2016, 2017; El
effect of an external load, were instead employed as a cost in Kim Makrini et al., 2018). This underlies the need to gather
et al. (2018a); Lorenzini et al. (2018); Kim et al. (2019, 2021b) to quantitative data giving insights about the mental
optimise the human body configuration in co-carrying or co- processing system.
manipulation activities according to human stability, shared A number of cross-sectional studies recorded the
workspace and task constraints. The cumulative effect of the physiological activity of the human during industrial-
overloading joint torque, i.e., overloading fatigue, was then mimicking tasks in laboratory settings and post-processed the
considered in a later work (Lorenzini et al., 2019) to trigger acquired biosignals to appraise how cognitive load develops

during HRC. Novak et al. (2011) assessed physical and cognitive motion directly from the input images of a stereo camera. In
load during haptic interaction with a robot by measuring ECG, particular, the online and quantitative framework examined
GSR, respiratory rate and peripheral skin temperature. The operators’ attention distribution, high activity periods and
combination of the last two measurements permitted body language while interacting with an industrial CoBot. The
estimating the mental cost in physically demanding tasks with degree of robot’s transparency and observability provided to the
haptic robots. Nevertheless, the proposed setup requires human worker had an influence on the development of cognitive
impractical equipment (i.e., a thermistor flow sensor), making workload. Moreover, changes in experienced mental effort
the technique hardly deployable in the industrial sector. Arai between human and automated assistance were mainly
et al. (2010) measured the mental strain in human-robot correlated to the operator’s familiarity with the technology
collaborative assembly tasks by recording the skin (Wurhofer et al., 2015). The transparency of robot behaviour
conductance and asking participants to fill a subjective was also considered by Dragan et al. (2013), Faber et al. (2016)
questionnaire. An increase in psychophysiological conditions and Gombolay et al. (2017) as criteria influencing cognitive
was found when a robot moved closer to a worker, with high ergonomics and included in the algorithm for ergonomic role
approaching speed and without notifying its motion in advance. allocation in hybrid industrial settings. In addition, Eimontaite
Analogous results were achieved by Kulić and Croft (2005, et al. (2019) explored the impact of graphical signage on
2007a,b) exploiting a more comprehensive range of participants physically collaborating with a semi-autonomous
physiological measurements (i.e., heart rate, corrugator muscle robot. According to data collected by questionnaires and facial
activity, GSR) and by Höcherl et al. (2017) and Bergman and van expression recognition, participants reported decreased anxiety
Zandbeek (2019) through subjective judgements. levels and negative attitudes toward the robot. On the other hand,
The interrelations between cognitive fatigue, operator sex the task-relevant signage supported task performance accuracy
and robot assistance level were also examined to optimise HRC rate and response time.
system designs with respect to task performance and user All of the studies reviewed here aim to appraise the impact of
experience. According to Hopko et al. (2021), the cognitive industrial CoBots and their actions on operators’ mental states in
effort, measured by HRV signals, had a negative impact on hybrid environments. The works of Lambrechts et al. (2021) and
task efficiency but did not change accuracy or precision. Gualtieri et al. (2021a) identified multiple cognitive ergonomics
Instead, the assistance through automation was subjectively variables in human-robot collaborative systems and underlined
perceived and rated in questionnaires as benefitting the importance of monitoring the human state and interpreting
performance in female subjects. Results of subjective nonverbal communication. Indeed, this information can be
judgements and secondary-task performance in Kaber and exploited to design frameworks capable of enhancing the
Endsley (2004) indicated that, when a more significant interaction between humans and robots by adapting online
percentage of primary task time was automated, operator the behaviour of the robotic teammates to operators’ needs.
perceptual resources were freed-up and monitoring The HRC solutions would build the foundations for
performance on the secondary task improved. In addition, improving cognitive ergonomics at work and mitigating the
Héraïz-Bekkis et al. (2020), Dragan et al. (2013), and Höcherl burden of work-related mental disorders worldwide.
et al. (2017) found that the feeling of perceived safety was Research following this principle is usually referred to as
enhanced when the robot motion was fluent and predictable. “affective robotics” (Braezeal et al., 2016), though current
More recent attention has focused on the provision of smooth approaches are mainly devoted to social or service robots.
trajectories to generate psychologically acceptable motions Moreover, the researchers are encountering difficulties due to the
without adding disturbances or uncomfortable sensations to multidimensional construct and high subjectivity of cognitive
the worker (Rojas et al., 2019, 2021). However, variations in processing and the typical human attitude of being ashamed and
perceived level of cognitive workload with the proposed robot’s concealing about psychological state. In Nicora et al. (2021), a
control strategies has not been tested yet. human-driven control architecture including a CoBot and an
Lately, three machine learning modalities were presented and interactive virtual Avatar is envisioned for promoting good
compared in Rajavenkatanarayanan et al. (2020) to extract relevant mental health. Preliminary attempts of introducing affective
features from ECG and GSR data, collected during robotic-assisted robotics in industrial settings were presented by Landi et al.
assembly tasks. The support vector machine (SVM) provided the (2018); Villani et al. (2018b, 2020). The proposed system
best accuracy of 92.85% in classifying the cognitive ergonomic risk estimated the user’s mental fatigue by analysing HRV and online
(i.e., low or high) potentially online. tuned the velocity of the slave, forbade hazardous manoeuvres or
With the aim to not introduce an additional source of provided assistive forces at the master interface. The interesting
discomfort for the worker (e.g., wearability constraints study by Messeri et al. (2021) offered a novel HRC paradigm where
introduced by biosensors) in manufacturing activities, the CoBot adapted its behaviour online based on the mutual
Lagomarsino et al. (2022a) monitored the mental effort and evaluation of the operator’s stress and productivity. More
stress level of human operators by detecting patterns in their specifically, the actual HRV parameters and cycle time were

TABLE 1 Physical ergonomics in industrial HRC: distribution of works among main monitoring systems and ergonomic assessment techniques.
Physical ergonomics in Systematic observations Direct measurements

human-robot
collaboration Pen-and-paper Model-based
Posture-based Comprehensive Motion and forces Electromyography

(EMG)
Kinodynamics Inertial-based Shafti et al. (2019), Van der Spaa Kim et al. (2018a), Kim et al.
Systems et al. (2020), Merlo et al. (2021) (2021b), Kim et al. (2019), Lorenzini
et al. (2018), Lorenzini et al. (2019),
Rapetti et al. (2019)
Camera-based Bestick et al. (2015), Busch et al. Mainprice et al. (2011), Mainprice
systems (2017), Busch et al. (2018), Pearce et al. (2012), Sisbot and Alami.
et al. (2018), Zanchettin et al. (2019), (2012), Marin et al. (2018),
El Makrini et al. (2019), Ferraguti Parastegari et al. (2017)
et al. (2020), Fusaro et al. (2021)
Force sensors Kim et al. (2018a), Kim et al.
(2021b), Kim et al. (2019), Lorenzini
et al. (2018), Lorenzini et al. (2019)
Physiological Traditional Peternel et al. (2017), Peternel

Biosensors et al. (2018), Peternel et al. (2019)
Portable
Biosensors
Noncontact
Biosensors
compared to the reference values retrieved by a game theory appraisals, from both the physical and cognitive/organisational
approach. The outcome was then exploited at each task points of view, with robotic strategies to mitigate the estimated
execution pipeline loop to vary the pace of interaction for hazards. This section first presents the main scientific
simultaneously mitigating the human cognitive workload and contribution of this paper, i.e., the identification of the main
maximising productivity. The trade-off between system gaps and thus potential research topics in the integration of
productivity and acceptable amount of human cognitive ergonomics principles within HRC frameworks. Next, the
workload in industrial tasks was also tackled by Lagomarsino operational implications are discussed and the practical
et al. (2022c). A multi-objective optimisation problem and an contribution is highlighted. Then, some insights for possible
online HRV-based decision-making algorithm were implemented future studies are provided and last, the limitations of this
to tune the total execution time and smoothness of the trajectories study are illustrated.
accomplished by the CoBot. This permits finding the most
appropriate pace of interaction for each specific user and online
adapting CoBot’s motion characteristics to fulfil changes in the 5.1 Identification of the gaps
individual needs.
Despite the growing concern for employees and employers Tables 1, 2 report the allocation of the leading research
worldwide in work-related stress and psychosocial risks, our themes according to the monitoring systems and the
careful investigation identified Lagomarsino et al. (2022a,c); ergonomic assessment method in the last 10 years (from
Landi et al. (2018); Messeri et al. (2021); Nicora et al. (2021); 2011 to 2022). As can be seen from Table 1, posture-based
Villani et al. (2018b, 2020) as the only studies which attempt to observational methods (such as RULA and REBA) are already
tackle the challenge of online assessing the cognitive demands widely studied to assess physical ergonomics in HRC and
and mitigating pressures at work. therefore known and consolidated from a research point of
view. Nevertheless, as already mentioned, such techniques are
limited to the kinematics of the workers’ actions. Just a couple of
5 Discussions and outlook researchers considered NIOSH/WISHA and SI, which take into
account also the task frequency, duration and the weight of the
The review shown in the previous section provides evidence involved objects. On the other hand, pen-and-paper
of the emerging and attractive interest in integrating ergonomic comprehensive techniques (i.e., including loads, action forces,

TABLE 2 Cognitive ergonomics in industrial HRC: distribution of works among main monitoring systems and ergonomic assessment techniques.
Cognitive ergonomics in Systematic observations Direct measurements

human-robot
collaboration Performance Behavioural assessment Physiological measures
measures
Electrocardiography (ECG) Galvanic
skin response (GSR)
Kinodynamics Inertial-based
systems
Camera-based Eimontaite et al. (2019),
systems Héraïz-Bekkis et al. (2020),
Lagomarsino et al. (2022a)
Force sensors
Physiological Traditional Messeri et al. (2021) Novak et al. (2011), Rajavenkatanarayanan Novak et al. (2011),
Biosensors et al. (2020), Hopko et al. (2021), Messeri Rajavenkatanarayanan et al.
et al. (2021) (2020)
Portable Lagomarsino et al. (2022c), Landi et al.
Biosensors (2018), Villani et al. (2018b), Villani et al.
(2020)
Noncontact
Biosensors
repetitions, etc., such as EAWS) have been already proposed in workload, and their (ii) fusion to obtain more reliable insights
an automatised version but have not found their practical into human fatigue and cognitive processes.
application so far as a policy for HRC. In general, however, Behavioural analysis is generally less investigated but,
the existing standard ergonomic tools present the great limitation according to the growth of paper production in recent years,
of not plenty considering the dynamics of the workers’activities, it represents an emerging research field in the context of
as discussed in Section 3. The authors who combined direct automatic detection of stress, frustration and anxiety. This is
measurements from motion and force sensors with human motivated by the advantages of low cost and operational ease of
models always proposed relatively simple metrics to address the assessment techniques. However, some disadvantages such as
subjects’ ergonomic demands (e.g., comfortable body vulnerability to motion and lack of burden-free calibration
configurations or decreased joint torque/fatigue). Indeed, solutions have not been completely addressed yet. In addition,
reduced-complexity human models with a limited number of due to their intrinsic properties, the authors identify noncontact
parameters allow fast identification processes and ensure minor sensors and cameras, maturing awareness and protection of
computational costs. privacy, as potential sensing systems that scientists should
By offering online monitoring of both human kinematics and concentrate and focus on to develop the next generation of
dynamics, this approach allows an exhaustive and off-the-shelf industrial CoBots taking into account workers’ ergonomics.
evaluation of the worker physical workload, while meeting the
requirements of the real factories. A few works exploit muscle
activity measured by EMG sensors to model physical fatigue in 5.2 Practical implications
hybrid environments. Despite their application limits in
industrial settings, biosensors and bodily signals could provide Crucial to the integration of what we would call “ergonomic
worthwhile information to enable the monitoring of the human- HRC frameworks” in real factories is the applicability and
robot pair. acceptability of the proposed solutions. The applicability is
Table 2 demonstrates that physiological indicators are related to the compatibility of the employed sensor systems
instead broadly explored for assessing cognitive workload, within the industrial environment, which are often noisy,
even if no well-established methodologies or commonly cluttered, and subject to frequent modifications. As discussed in
accepted metrics for the cognitive workload exist in the Section 3, to monitor the human state, both wearable and external
literature. Review results reveal a growing interest in devices are available, which can be preferable depending on the tasks
unobtrusive sensing and wearable devices. The latter permits to be executed and the work place characteristics and requirements.
the (i) simultaneous monitoring of multiple physical and In general, the sensor technologies should be selected to ensure a
physiological signals, sensitive to distinctive aspects of reliable monitoring of the workers’ state but also maximise their

comfort without hindering daily activity. Indeed, the acceptability is results show excellent potentials in mitigating work-related
related to workers’ opinion. Since they will firsthand use and biomechanical and cognitive workload without introducing
experience the proposed technologies, their approval is new occupational safety and health hazards. From a physical
fundamental for an effective integration. Hence, parallel to the point of view, the existing literature offers a variety of ergonomic
development of HRC strategies, their evaluation with real metrics that became well-established tools in the industrial
workers in real factories is a key requirement. To date, several environment. In parallel, advanced modelling and estimation
questionnaires are available for this purpose, e.g., the system algorithms can make human kinodynamic variables available
usability scale (SUS) (Brooke, 1996), Borg scale (Borg et al., online, but most of the underlying techniques are still confined in
1985), Positive and Negative Affect Schedule (PANAS) laboratory settings. As such, the most significant and prominent
(Thompson, 2007) and also the NASA-TLX (Hart and research themes for physical ergonomics are the integration and
Staveland, 1988; Hart, 2006), but should be updated to keep transfer of these methods in the workplace. To this end, the
apace with the technological advancements. Even the most potential of wearable sensor technologies should be exploited to
efficient and ergonomic equipment, without users’ approval, maximise users’ comfort without hindering workers’ daily
become meaningless. We believe that this review paper can be activity. Moreover, future research should proceed with
useful not only for researchers willing to discover new research implementing role allocation strategies in hybrid
themes but also to business executives and employee representatives environments and online planning and adaptation of robot
to get an overview about the state-of-the-art of HRC solutions to movements. All those solutions fall within the cutting-edge
address human factors and ergonomics. A continuous dialogue of principle of designing human-centred workstations supported
robotics (but non only) researchers with these entities to gather by automation and could build the foundation for a more
information about applicability and acceptability as well as make inclusive industrial environment. Thanks to the introduction
them aware of the available possibilities in terms of technology is of ergonomics assessment in the control loop, the robot
fundamental for advancements in this field. behaviour could be adapted to workers’ physical condition
On the other hand, to work along the right lines, robotics and characteristics (e.g., age, gender, dominant and vulnerable
researchers should receive counselling and support from limb, disabilities, fatigue) and enhance specific skills, mitigate
ergonomics specialist and practitioners to ensure the correct risks, fulfil required capabilities and fight inequalities.
use and applications of ergonomics principles. The exchange of To date, research on ergonomics in the industrial sector
information between these two domains is currently too scarse mainly aims at mitigating the worker’s fatigue and discomfort
while it would enhance and fasten the integration of human from a physical point of view. Future research should entail
ergonomics within HRC. By providing simultaneously an cognitive ergonomics variables, whose implications are still too
overview of the current tools to assess human ergonomics often undervalued (HSE, 2020). Industrial collaborative
along with the available technologies to monitor the human technologies provide unique opportunities, but they may
state, up to their combination in the design of HRC frameworks, perilously increase operators’ mental demand when
we hope to highlight the importance of the constant interplay inadequately handled and result in adverse health and safety
between the ergonomics and robotics communities. hazards. Assuring the acceptability of robotic systems from
human workers and guaranteeing perceived safety are the first
steps for a successful workplaces’ digitalisation. However, the
5.3 Future research trends scientific and industrial communities still need to be provided
with a well-structured and validated set of models and metrics
Despite the growing enthusiasm to understand the for the cognitive workload. Then, in the near future, researchers
development of MSDs and the multidimensional construct of should concentrate on developing reliable methodologies of the
the mental workload, the study of human ergonomics in mental demand inflicted by industrial tasks. This could be
collaborative robotic workstations is relatively a new topic, exploited afterwards in new research lines aimed to
still looking for practical solutions. Most of the studies on maximise workers’ efficiency and workstation productivity
focus on the impact of industrial CoBots and their actions on and facilitate the adoption of CoBots in real-world industrial
operators’ physical and mental states. The literature collection environments.
and analysis presented in this article shows that the actual
research tends to rely on elaborate and time-consuming data
post-processing. Consequently, available tools can be used almost 5.4 Limitations of this study
exclusively by experts or merely provide subjective and offline
insights about human ergonomics, inhibiting their applicability The review covered four out of five sub-parts identified by
in real-world environments. Preliminary attempts to gather Hendrick (1998); Hendrick and Kleiner (2002) within the field of
online data and accordingly adapt robot behaviour are human factors/ergonomics, i.e., human-machine interface, human-
investigated. Although limited to laboratory experiments, environment interface, human-software interface, human-job

interface technologies. These technologies, which are primarily Author contributions

focused at the individual level, are often referred to in the
literature as “microergonomics.” The exclusion of the structural MLo, MLa, LF, SG, and AA contributed to the conception
dimensions of work systems, i.e., human-organisation interface and design of the study. MLo, MLa, LF, and SG performed the
technology or “macroergonomics,” was mainly motivated by the literature review and wrote the first draft of the manuscript. MLo,
desire to investigate ergonomic metrics that could be exploited to MLa, LF, SG, and AA contributed to manuscript revision and
drive and adapt the robot behaviour and potentially improve approved the submitted version. AA supervised the work and
ergonomics in hybrid environments. The boundaries of our founded the research project.
analysis permit to comprehensively investigate and discuss
existing ergonomics assessment tools and available monitoring
devices. Nonetheless, systematic macroergonomic methodologies Funding
provide a larger perspective of the overall work system and could
increase the likelihood of the microergonomic interventions This study was supported in part by the European Research
presented in this review having a more significant impact and Council (ERC) starting grant Ergo-Lean (GA 850932), and by the
effectiveness. European Union’s Horizon 2020 research and innovation
programme SOPHIA (GA 871237).
6 Conclusion
Conflict of interest
The goal of this article was to provide an overview of the
current state-of-the-art in ergonomic HRC in industrial settings. The authors declare that the research was conducted in the
Ergonomic assessment methodologies and available monitoring absence of any commercial or financial relationships that could be
technologies to online adapt robot control strategy according to construed as a potential conflict of interest.
workers’ distress and needs were investigated, and the most
promising research themes were highlighted. Despite the
booming attention in physical and cognitive ergonomic HRC, Publisher’s note
several challenges are still waiting to be solved. In particular,
when the technologies for ergonomics monitoring and HRC will All claims expressed in this article are solely those of the
reach a more mature level, the challenges to be addressed include authors and do not necessarily represent those of their
the cost-effectiveness, the level of expertise needed to implement affiliated organizations, or those of the publisher, the
and maintain them, and the multi-person examination capacity. editors and the reviewers. Any product that may be
These challenges are not only limited to the technical aspects but evaluated in this article, or claim that may be made by
also to the regulatory ones, such as privacy issues when it comes its manufacturer, is not guaranteed or endorsed by the
to monitoring of humans. publisher.
References
ACGIH (1981). “TLVs: Threshold limit values for chemical substances and physical Arjmand, N., Gagnon, D., Plamondon, A., Shirazi-Adl, A., and Lariviere, C. (2009).
substances in the workroom environment with intended changes for 1981,” in American Comparison of trunk muscle forces and spinal loads estimated by two biomechanical
Conference of Governmental Industrial Hygienists, Cincinnati, OH, USA. models. Clin. Biomech. 24, 533–541. doi:10.1016/j.clinbiomech.2009.05.008
Ahlstrom, U., and Friedman-Berg, F. J. (2006). Using eye movement activity as a Ayusawa, K., Venture, G., and Nakamura, Y. (2014). Identifiability and
correlate of cognitive workload. Int. J. Industrial Ergonomics 36, 623–636. doi:10. identification of inertial parameters using the underactuated base-link dynamics
1016/j.ergon.2006.04.002 for legged multibody systems. Int. J. Robotics Res. 33, 446–468. doi:10.1177/
0278364913495932
Aigrain, J., Dubuisson, S., Detyniecki, M., and Chetouani, M. (2015). “Person-
specific behavioural features for automatic stress detection,” in Proceedings of Bailey, S. W., and Bodenheimer, B. (2012). “A comparison of motion capture data
International Conference and Workshops on Automatic Face and Gesture recorded from a vicon system and a microsoft kinect sensor,” in Proceedings of
Recognition (FG), Ljubljana, Slovenia (IEEE). Symposium on Applied Perception (SAP), Los Angeles, CA, USA (New York, NY:
ACM Press), 121.
Al-Shargie, F., Tang, T. B., Badruddin, N., and Kiguchi, M. (2016). “Mental stress
quantification using EEG signals,” in Proceedings of International Conference for Bamberg, S., Benbasat, A., Scarborough, D., Krebs, D., and Paradiso, J. (2008).
Innovation in Biomedical Engineering and Life Sciences (ICIBEL), Singapore Gait analysis using a shoe-integrated wireless sensor system. IEEE Trans. Inf.
(Springer), 15–19. Technol. Biomed. 12, 413–423. doi:10.1109/titb.2007.899493
Andreas, G.-W. J., and Johanssons, E. (2018). Observational methods for assessing Berg, J., and Lu, S. (2020). Review of interfaces for industrial human-robot
ergonomic risks for work-related musculoskeletal disorders. a scoping review. Rev. Cienc. interaction. Curr. Robot. Rep. 1, 27–34. doi:10.1007/s43154-020-00005-6
salud 16, 8–38. doi:10.12804/revistas.urosario.edu.co/revsalud/a.6840
Bergman, M., and van Zandbeek, M. (2019). Close encounters of the fifth kind?
Arai, T., Kato, R., and Fujita, M. (2010). Assessment of operator stress induced by Affective impact of speed and distance of a collaborative industrial robot on
robot collaboration in assembly. CIRP Ann. 59, 5–8. doi:10.1016/j.cirp.2010.03.043 humans. Hum. Friendly Robot. 7, 127–137.

Bestick, A., Bajcsy, R., and Dragan, A. D. (2016). “Implicitly assisting humans to Castro, P. R., Högberg, D., Ramsen, H., Bjursten, J., and Hanson, L. (2018).
choose good grasps in robot to human handovers,” in Proceedings of International “Virtual simulation of human-robot collaboration workstations,” in Proceedings of
Symposium on Experimental Robotics, Nagasaki, Japan (Springer), 341–354. Congress of the International Ergonomics Association (IEA), Florence, Italy
(Springer), 250–261.
Bestick, A. M., Burden, S. A., Willits, G., Naikal, N., Sastry, S. S., and Bajcsy, R.
(2015). “Personalized kinematics for human-robot collaborative manipulation,” in Chao, E. Y., Armiger, R., Yoshida, H., Lim, J., and Haraguchi, N. (2007). Virtual
Proceedings of International Conference on Intelligent Robots and Systems (IROS), interactive musculoskeletal system (vims) in orthopaedic research, education and
Hamburg, Germany (IEEE), 1037–1044. clinical patient care. J. Orthop. Surg. Res. 2, 2. doi:10.1186/1749-799x-2-2
Bestick, A., Pandya, R., Bajcsy, R., and Dragan, A. D. (2018). “Learning human Chatzitofis, A., Zarpalas, D., Daras, P., and Kollias, S. (2021). DeMoCap: Low-cost
ergonomic preferences for handovers,” in Proceedings of International Conference marker-based motion capture. Int. J. Comput. Vis. 129, 3338–3366. doi:10.1007/
on Robotics and Automation (ICRA), Brisbane, Australia (IEEE), 3257–3264. s11263-021-01526-z
Bhargava, L., Pandy, M., and Anderson, F. (2004). A phenomenological model for Christmansson, M. (1994). “The hama-method: A new method for analysis of
estimating metabolic energy consumption in muscle contraction. J. biomechanics upper limb movements and risk for work-related musculoskeletal disorders,” in
37, 81–88. doi:10.1016/s0021-9290(03)00239-2 Proceedings of Congress of the International Ergonomics Association/Human
Factors Association of Canada, Toronto, Canada.
Bi, Z., Luo, C., Miao, Z., Zhang, B., Zhang, W., and Wang, L. (2021). Safety
assurance mechanisms of collaborative robotic systems in manufacturing. Robotics Colim, A., Morgado, R., Carneiro, P., Costa, N., Faria, C., Sousa, N., et al. (2021).
Computer-Integrated Manuf. 67, 102022. doi:10.1016/j.rcim.2020.102022 Lean manufacturing and ergonomics integration: Defining productivity and
wellbeing indicators in a human–robot workstation. Sustainability 13, 1931.
Bigos, S., Battié, M. C., Spengler, D., Fisher, L., Fordyce, W., Hansson, T., et al.
doi:10.3390/su13041931
(1991). A prospective study of work perceptions and psychosocial factors affecting
the report of back injury. Spine 16, 1–6. doi:10.1097/00007632-199101000-00001 Colyer, S. L., Evans, M., Cosker, D. P., and Salo, A. I. T. (2018). A review of the
evolution of vision-based motion analysis and the integration of advanced
Bisogni, C., Hao, F., Loia, V., and Narducci, F. (2022). Drowsiness detection in the
computer vision methods towards developing a markerless system. Sports Med.
era of industry 4.0: Are we ready. IEEE Trans. Ind. Inf. 1. doi:10.1109/TII.2022.
Open 4, 24–15. doi:10.1186/s40798-018-0139-y
3173004
Corlett, E. N., and Bishop, R. (1976). A technique for assessing postural
Bongiovanni, A., De Luca, A., Gava, L., Grassi, L., Lagomarsino, M., Lapolla, M.,
discomfort. Ergonomics 19, 175–182. doi:10.1080/00140137608931530
et al. (2022). Gestural and touchscreen interaction for human-robot collaboration: A
comparative study. arXiv preprint arXiv:2207.03783. Corlett, E. N., Madeley†, * S. J., and Manenica‡, I. (1979). Posture targeting: A
technique for recording working postures. Ergonomics 22, 357–366. doi:10.1080/
Borg, G., Ljunggren, G., and Ceci, R. (1985). The increase of perceived exertion, aches
00140137908924619
and pain in the legs, heart rate and blood lactate during exercise on a bicycle ergometer.
Eur. J. Appl. Physiol. Occup. Physiol. 54, 343–349. doi:10.1007/bf02337176 Cox, T., and Mackay, C. (1985). The measurement of self-reported stress and
arousal. Br. J. Psychol. 76, 183–186. doi:10.1111/j.2044-8295.1985.tb01941.x
Bortolini, M., Faccio, M., Gamberi, M., and Pilati, F. (2020). Motion analysis
system (MAS) for production and ergonomics assessment in the manufacturing Coyne, J., and Sibley, C. (2016). “Investigating the use of two low cost eye tracking
processes. Comput. Ind. Eng. 139, 105485. doi:10.1016/j.cie.2018.10.046 systems for detecting pupillary response to changes in mental workload,” in
Proceedings of Human Factors and Ergonomics Society (HFES) Annual
Bortolini, M., Gamberi, M., Pilati, F., and Regattieri, A. (2018). Automatic
Meeting, Washington, DC, USA (Los Angeles, CA: SAGE Publications), 37–41.60
assessment of the ergonomic risk for manual manufacturing and assembly
activities through optical motion capture technology. Procedia CIRP 72, 81–86. Critchley, H. D. (2002). Review: Electrodermal responses: What happens in the
doi:10.1016/j.procir.2018.03.198 brain. Neuroscientist 8, 132–142. doi:10.1177/107385840200800209
Braezeal, C., Takanishi, A., and Kobayashi, T. (2016). Handbook of robotics. Damgrave, R. G. J., and Lutters, D. (2009). “The drift of the xsens moven motion
Springer. chap. 58: Social Robots that Interact with People. 1349–1369. capturing suit during common movements in a working environment,” in
Proceedings of CIRP Design Conference, Cranfield, UK (Cranfield, United
Brooke, J. (1996). Sus-a quick and dirty usability scale. Usability Eval. industry
Kingdom: Cranfield University Press), 277–292.
189, 4–7.
Damsgaard, M., Rasmussen, J., Christensen, S., Surma, E., and De Zee, M. (2006).
Brosque, C., Galbally, E., Khatib, O., and Fischer, M. (2020). “Human-robot
Analysis of musculoskeletal systems in the anybody modeling system. Simul. Model.
collaboration in construction: Opportunities and challenges,” in 2020 International
Pract. Theory 14, 1100–1111. doi:10.1016/j.simpat.2006.09.001
Congress on Human-Computer Interaction, Optimization and Robotic
Applications (HORA IEEE), 1–8. David, G. C. (2005). Ergonomic methods for assessing exposure to risk factors for
work-related musculoskeletal disorders. Occup. Med. 55, 190–199. doi:10.1093/
Buchanan, T. S., Lloyd, D. G., Manal, K., and Besier, T. F. (2004).
occmed/kqi082
Neuromusculoskeletal modeling: Estimation of muscle forces and joint moments
and movements from measurements of neural command. J. Appl. biomechanics 20, de Kok, J., Vroonhof, P., Snijders, J., Roullis, G., Clarke, M., Peereboom, K., et al.
367–395. doi:10.1123/jab.20.4.367 (2019). Work-related musculoskeletal disorders: Prevalence, costs and demographics
in the EU. Tech. rep. Bilbao, Spain: EU-OSHA European Agency for Safety and
Burdorf, A. (1992). Exposure assessment of risk factors for disorders of the back
Health at Work.
in occupational epidemiology. Scand. J. Work Environ. Health 18, 1–9. doi:10.5271/
sjweh.1615 De Luca, C., Gilmore, L., Kuznetsov, M., and Roy, S. (2010). Filtering the surface
emg signal: Movement artifact and baseline noise contamination. J. biomechanics
Burdorf, A., and Laan, J. (1991). Comparison of methods for the assessment of
43, 1573–1579. doi:10.1016/j.jbiomech.2010.01.027
postural load on the back. Scand. J. Work Environ. Health 17, 425–429. doi:10.5271/
sjweh.1679 de Santos Sierra, A., Avila, C. S., del Pozo, G. B., and Casanova, J. G. (2011). “Stress
detection by means of stress physiological template,” in Proceedings of World Congress
Busch, B., Maeda, G., Mollard, Y., Demangeat, M., and Lopes, M. (2017).
on Nature and Biologically Inspired Computing, Salamanca, Spain (IEEE).
“Postural optimization for an ergonomic human-robot interaction,” in
Proceedings of International Conference on Intelligent Robots and Systems del Rio Vilas, D., Longo, F., and Monteil, N. R. (2013). A general framework for
(IROS), Vancouver, Canada (IEEE), 2778–2785. the manufacturing workstation design optimization: A combined ergonomic and
operational approach. Simulation 89, 306–329. doi:10.1177/0037549712462862
Busch, B., Toussaint, M., and Lopes, M. (2018). “Planning ergonomic sequences
of actions in human-robot interaction,” in Proceedings of International Conference Delliaux, S., Delaforge, A., Deharo, J.-C., and Chaumet, G. (2019). Mental
on Robotics and Automation (ICRA), Brisbane, Australia (IEEE), 1916–1923. workload alters heart rate variability, lowering non-linear dynamics. Front.
Physiol. 10, 565. doi:10.3389/fphys.2019.00565
Cao, Z., Hidalgo, G., Simon, T., Wei, S.-E., and Sheikh, Y. (2021). OpenPose:
Realtime multi-person 2d pose estimation using part affinity fields. IEEE Trans. Delp, S., Anderson, F., Arnold, A., Loan, P., Habib, A., John, C., et al. (2007).
Pattern Anal. Mach. Intell. 43, 172–186. doi:10.1109/tpami.2019.2929257 Opensim: Open-source software to create and analyze dynamic simulations of
movement. IEEE Trans. Biomed. Eng. 54, 1940–1950. doi:10.1109/tbme.2007.
Caputo, F., Greco, A., D‘Amato, E., Notaro, I., and Spada, S. (2018). IMU-based
901024
motion capture wearable system for ergonomic assessment in industrial
environment. Adv. Intelligent Syst. Comput. 795, 215–225. Dickinson, C., Campion, K., Foster, A., Newman, S., O’Rourke, A., and Thomas,
P. (1992). Questionnaire development: An examination of the nordic
Carrasco, G. A., and de Kar, L. D. V. (2003). Neuroendocrine pharmacology of
musculoskeletal questionnaire. Appl. Ergon. 23, 197–201. doi:10.1016/0003-
stress. Eur. J. Pharmacol. 463, 235–272. doi:10.1016/s0014-2999(03)01285-8
6870(92)90225-k
Castro, A., Silva, F., and Santos, V. (2021). Trends of human-robot collaboration
Dimitropoulos, N., Togias, T., Zacharaki, N., Michalos, G., and Makris, S. (2021).
in industry contexts: Handover, learning, and metrics. Sensors 21, 4113. doi:10.
Seamless human–robot collaborative assembly using artificial intelligence and
3390/s21124113
wearable devices. Appl. Sci. 11, 5699. doi:10.3390/app11125699

Douwes, M., and Kraker, H. d. (2009). “Hand arm risk assessment method systems,” in Proceedings of Congress of the International Ergonomics Association
(HARM), a new practical tool,” in Proceedings of World Congress on Ergonomics, (virtual (Springer), 266–273.
Beijing, China: (Geneva, Switzerland: International Ergonomics Association
Gualtieri, L., Palomba, I., Merati, F. A., Rauch, E., and Vidoni, R. (2020). Design of
(IEA)), 9–14.
human-centered collaborative assembly workstations for the improvement of
Dragan, A. D., Lee, K. C., and Srinivasa, S. S. (2013). “Legibility and predictability operators’ physical ergonomics and production efficiency: A case study.
of robot motion,” in Proceedings of International Conference on Human-Robot Sustainability 12, 3606. doi:10.3390/su12093606
Interaction (HRI), Tokyo, Japan (IEEE), 301–308. doi:10.1109/HRI.2013.6483603
Gualtieri, L., Rauch, E., and Vidoni, R. (2021b). Emerging research fields in safety
Drury, C., and Coury, B. (1982). A methodology for chair evaluation. Appl. Ergon. and ergonomics in industrial collaborative robotics: A systematic literature review.
13, 195–202. doi:10.1016/0003-6870(82)90006-0 Robotics Computer-Integrated Manuf. 67, 101998. doi:10.1016/j.rcim.2020.101998
Durantin, G., Gagnon, J.-F., Tremblay, S., and Dehais, F. (2014). Using near Haggag, H., Hossny, M., Nahavandi, S., and Creighton, D. (2013). “Real time
infrared spectroscopy and heart rate variability to detect mental overload. Behav. ergonomic assessment for assembly operations using kinect,” in Proceedings of
Brain Res. 259, 16–23. doi:10.1016/j.bbr.2013.10.042 International Conference on Computer Modelling and Simulation, Rome, Italy
(IEEE), 495–500.
Eimontaite, I., Gwilt, I., Cameron, D., Aitken, J. M., Rolph, J., Mokaram, S., et al.
(2019). Language-free graphical signage improves human performance and reduces Han, H. J., Labbaf, S., Borelli, J. L., Dutt, N., and Rahmani, A. M. (2020). Objective
anxiety when working collaboratively with robots. Int. J. Adv. Manuf. Technol. 100, stress monitoring based on wearable sensors in everyday settings. J. Med. Eng.
55–73. doi:10.1007/s00170-018-2625-2 Technol. 44, 177–189. doi:10.1080/03091902.2020.1759707
El Makrini, I., Elprama, S. A., Van den Bergh, J., Vanderborght, B., Knevels, A.-J., Hart, S. G. (2006). “Nasa-task load index (NASA-TLX) 20 years later,” in
Jewell, C. I., et al. (2018). Working with walt: How a cobot was developed and Proceedings of Human Factors and Ergonomics Society (HFES) Annual
inserted on an auto assembly line 25, 51–58. Meeting (Los Angeles, CA: SAGE Publications), 904–908.
El Makrini, I., Merckaert, K., De Winter, J., Lefeber, D., and Vanderborght, B. Hart, S. G., and Staveland, L. E. (1988). Development of NASA-TLX (task load
(2019). Task allocation for improved ergonomics in human-robot collaborative index): Results of empirical and theoretical research. Adv. Psychol., 139–183. doi:10.
assembly. Interact. Stud. 20, 102–133. doi:10.1075/is.18018.mak 1016/s0166-4115(08)62386-9
Elprama, S., El Makrini, I., Vanderborght, B., and Jacobs, A. (2016). “Acceptance Hassard, J., Teoh, K., Cox, T., Dew, P., Cosmar, M., Gründler, R., et al. (2014).
of collaborative robots by factory workers: A pilot study on the role of social cues of Calculating the cost of work-related stress and psychosocial risks. Tech. rep. Bilbao,
anthropomorphic robots,” in Proceedings of International Conference on Robot Spain: EU-OSHA European Agency for Safety and Health at Work.
and Human Interactive Communication (RO-MAN), New York, USA (IEEE).
Hendrick, H., and Kleiner, B. (2002). Macroergonomics: Theory, methods, and
Elprama, S., Jewell, C., Jacobs, A., El Makrini, I., and Vanderborght, B. (2017). applications. Boca Raton, FL: CRC Press.
“Attitudes of factory workers towards industrial and collaborative robots,” in
Hendrick, H. (1998). Macroergonomics: Theory, methods, and applications. Adv.
Proceedings of International Conference on Human-Robot Interaction, Vienna,
Occup. ergonomics Saf. 2, 26–34.
Austria (ACM/IEEE), 113–114.
Henelius, A., Hirvonen, K., Holm, A., Korpela, J., and Muller, K. (2009). “Mental
Faber, M., Kuz, S., Mertens, A., and Schlick, C. M. (2016). Model-based
workload classification using heart rate metrics,” in Proceedings of International
evaluation of cooperative assembly processes in human-robot collaboration.
Conference of the IEEE Engineering in Medicine and Biology Society (EMBS),
Adv. Ergonomics Manuf. Manag. Enterp. Future. Adv. Intelligent Syst. Comput.
Minneapolis, MN, USA (IEEE), 1836–1839.
490, 101–112.
Héraïz-Bekkis, D., Ganesh, G., Yoshida, E., and Yamanobe, N. (2020). “Robot
Farina, D., and Merletti, R. (2001). A novel approach for precise simulation of the
movement uncertainty determines human discomfort in Co-worker scenarios,” in
emg signal detected by surface electrodes. IEEE Trans. Biomed. Eng. 48, 637–646.
Proceedings of International Conference on Control, Automation and Robotics
doi:10.1109/10.923782
(ICCAR, Singapore (IEEE), 59–66.
Ferraguti, F., Villa, R., Landi, C. T., Zanchettin, A. M., Rocco, P., and Secchi, C.
Herman, I. (2007). Physics of the human body. Springer. chap. 1: Terminology, the
(2020). A unified architecture for physical and ergonomic human–robot
standard human, and scaling.
collaboration. Robotica 38, 669–683. doi:10.1017/s026357471900095x
Heydaryan, S., Suaza Bedolla, J., and Belingardi, G. (2018). Safety design and
Forster, E. (2004). Predicting muscle forces in the human lower limb during
development of a human-robot collaboration assembly process in the automotive
locomotion. Ph.D. thesis. Ulm, Germany: Universität Ulm.
industry. Appl. Sci. 8, 344. doi:10.3390/app8030344
Fraysse, F., Dumas, R., Cheze, L., and Wang, X. (2009). Comparison of global
Hignett, S., and McAtamney, L. (2000). Rapid entire body assessment (reba).
and joint-to-joint methods for estimating the hip joint load and the muscle forces
Appl. Ergon. 31, 201–205. doi:10.1016/s0003-6870(99)00039-3
during walking. J. biomechanics 42, 2357–2362. doi:10.1016/j.jbiomech.2009.
06.056 Hill, A. (1938). “The heat of shortening and the dynamic constants of muscle,” in
Proceedings of the Royal Society of London. Series B-Biological Sciences, London,
Fusaro, F., Lamon, E., De Momi, E., and Ajoudani, A. (2021). “A human-aware
United Kingdom (London, United Kingdom: The Royal Society London), 136–195.
method to plan complex cooperative and autonomous tasks using behavior trees,”
in Proceedings of International Conference on Humanoid Robots (Humanoids), Hinde, K., White, G., and Armstrong, N. (2021). Wearable devices suitable for
Munich, Germany (IEEE), 522–529. monitoring twenty four hour heart rate variability in military populations. Sensors
21, 1061. doi:10.3390/s21041061
Garg, A., Moore, J. S., and Kapellusch, J. M. (2017). The revised strain index: An
improved upper extremity exposure assessment model. Ergonomics 60, 912–922. Höcherl, J., Wrede, B., and Schlegl, T. (2017). “Motion analysis of human-human
doi:10.1080/00140139.2016.1237678 and human-robot cooperation during industrial assembly tasks,” in Proceedings of
International Conference on Human Agent Interaction (HAI), New York, NY, USA
Gholami, S., Lorenzini, M., De Momi, E., and Ajoudani, A. (2022). Quantitative
(New York, NY: ACM Press), 425–429.
physical ergonomics assessment of teleoperation interfaces. IEEE Trans. Hum.
Mach. Syst. 52, 169–180. doi:10.1109/thms.2022.3149167 Hoffman, S. G., Reed, M. P., and Chaffin, D. B. (2007). Predicting force-exertion
postures from task variables. SAE Tech. Pap.
Giakoumis, D., Drosou, A., Cipresso, P., Tzovaras, D., Hassapis, G., Gaggioli, A.,
et al. (2012). Using activity-related behavioural features towards more effective Hopko, S., Khurana, R., Mehta, R. K., and Pagilla, P. R. (2021). Effect of cognitive
automatic stress detection. PLoS ONE 7, e43571–16. doi:10.1371/journal.pone. fatigue, operator sex, and robot assistance on task performance metrics, workload,
0043571 and situation awareness in human-robot collaboration. IEEE Robot. Autom. Lett. 6,
3049–3056. doi:10.1109/lra.2021.3062787
Glowinski, D., Dael, N., Camurri, A., Volpe, G., Mortillaro, M., and Scherer, K.
(2011). Toward a minimal representation of affective gestures. IEEE Trans. Affect. HSE (2020). Work-related stress, anxiety or depression statistics in Great Britain,
Comput. 2, 106–118. doi:10.1109/t-affc.2011.7 2020. Tech. rep. Bootle, United Kingdom: Health and Safety Executive HSE.
Gombolay, M., Bair, A., Huang, C., and Shah, J. (2017). Computational design of Hu, H., Cao, Z., Yang, X., Xiong, H., and Lou, Y. (2021). Performance evaluation
mixed-initiative human–robot teaming that considers human factors: Situational of optical motion capture sensors for assembly motion capturing. IEEE Access 9,
awareness, workload, and workflow preferences. Int. J. Robotics Res. 36, 597–617. 61444–61454. doi:10.1109/access.2021.3074260
doi:10.1177/0278364916688255
Hughes, A. M., Hancock, G. M., Marlow, S. L., Stowers, K., and Salas, E. (2019).
Grassmann, M., Vlemincx, E., von Leupoldt, A., Mittelstädt, J. M., and den Bergh, Cardiac measures of cognitive workload: A meta-analysis. Hum. Factors 61,
O. V. (2016). Respiratory changes in response to cognitive load: A systematic 393–414. doi:10.1177/0018720819830553
review. Neural Plast., 1–16. doi:10.1155/2016/8146809
James, S. L., Abate, D., Abate, K. H., Abay, S. M., Abbafati, C., Abbasi, N., et al.
Gualtieri, L., Fraboni, F., Marchi, M. D., and Rauch, E. (2021a). “Evaluation of (2018). Global, regional, and national incidence, prevalence, and years lived with
variables of cognitive ergonomics in industrial human-robot collaborative assembly disability for 354 diseases and injuries for 195 countries and territories, 1990–2017:

A systematic analysis for the global burden of disease study 2017. Lancet 392, reality human-robot interaction,” in Proceedings of International Conference on
1789–1858. doi:10.1016/S0140-6736(18)32279-7 Intelligent Robots and Systems (IROS), Madrid, Spain (IEEE), 1–9.
Jebelli, H., Choi, B., and Lee, S. (2019). Application of wearable biosensors to Kubicek, B., Paškvan, M., Prem, R., Schöllbauer, J., Till, M., Cabrita, J., et al.
construction sites. i: Assessing workers’ stress. J. Constr. Eng. Manag. 145, (2019). Working conditions and workers’ health. Tech. Rep. Eurofound.
04019079. doi:10.1061/(asce)co.1943-7862.0001729
Kulić, D., and Croft, E. A. (2007b). Affective state estimation for human–robot
Kaber, D. B., and Endsley, M. R. (2004). The effects of level of automation and interaction. IEEE Trans. Robot. 23, 991–1000. doi:10.1109/TRO.2007.904899
adaptive automation on human performance, situation awareness and workload in
Kulić, D., and Croft, E. (2005). “Anxiety detection during human-robot
a dynamic control task. Theor. Issues Ergonomics Sci. 5, 113–153. doi:10.1080/
interaction,” in Proceedings of International Conference on Intelligent Robots
1463922021000054335
and Systems (IROS, Alberta, Canada (IEEE).
Kaber, D. B., and Riley, J. M. (1999). Adaptive automation of a dynamic control
Kulić, D., and Croft, E. (2007a). Physiological and subjective responses to
task based on secondary task workload measurement. Int. J. Cognitive Ergonomics 3,
articulated robot motion. Robotica 25, 13–27. doi:10.1017/s0263574706002955
169–187. doi:10.1207/s15327566ijce0303_1
Kumar, D., Rudolph, K., and Manal, K. (2012). Emg-driven modeling approach to
Kadir, B. A., Broberg, O., and da Conceicao, C. S. (2019). Current research and
muscle force and joint load estimations: Case study in knee osteoarthritis.
future perspectives on human factors and ergonomics in industry 4.0. Comput.
J. Orthop. Res. 30, 377–383. doi:10.1002/jor.21544
Industrial Eng. 137, 106004. doi:10.1016/j.cie.2019.106004
Kumar, S., Savur, C., and Sahin, F. (2020). Survey of human–robot collaboration
Kanko, R. M., Laende, E. K., Davis, E. M., Selbie, W. S., and Deluzio, K. J. (2021).
in industrial settings: Awareness, intelligence, and compliance. IEEE Trans. Syst.
Concurrent assessment of gait kinematics using marker-based and markerless
Man. Cybern. Syst. 51, 280–297. doi:10.1109/tsmc.2020.3041231
motion capture. J. Biomechanics 127, 110665. doi:10.1016/j.jbiomech.2021.110665
Kuorinka, I., Jonsson, B., Kilbom, A., Vinterberg, H., Biering-Sørensen, F.,
Karg, M., Samadani, A.-A., Gorbet, R., Kuhnlenz, K., Hoey, J., and Kulić, D.
Andersson, G., et al. (1987). Standardised nordic questionnaires for the analysis
(2013). Body movements for affective expression: A survey of automatic recognition
of musculoskeletal symptoms. Appl. Ergon. 18, 233–237. doi:10.1016/0003-
and generation. IEEE Trans. Affect. Comput. 4, 341–359. doi:10.1109/t-affc.2013.29
6870(87)90010-x
Karhu, O., Kansi, P., and Kuorinka, I. (1977). Correcting working postures in
Kyriakou, K., Resch, B., Sagl, G., Petutschnig, A., Werner, C., Niederseer, D., et al.
industry: A practical method for analysis. Appl. Ergon. 8, 199–201. doi:10.1016/
(2019). Detecting moments of stress from measurements of wearable physiological
0003-6870(77)90164-8
sensors. Sensors 19, 3805. doi:10.3390/s19173805
Katayama, M., and Hasuura, H. (2003). “Optimization principle determines
Lagomarsino, M., Lorenzini, M., Balatti, P., De Momi, E., and Ajoudani, A.
human arm postures and comfort,” in Proceedings of Annual Conference Program
(2022a). Pick the right Co-worker: Online assessment of cognitive ergonomics in
and Abstracts SICE Annual Conference, Fukui, Japan (Tokyo, Japan: The Society of
human-robot collaborative assembly. IEEE Trans. Cogn. Dev. Syst., 1. doi:10.1109/
Instrument and Control Engineers SICE), 47.
tcds.2022.3182811
Kee, D., and Karwowski, W. (2001). Luba: An assessment technique for postural
Lagomarsino, M., Lorenzini, M., De Momi, E., and Ajoudani, A. (2022b). An online
loading on the upper body based on joint motion discomfort and maximum holding
framework for cognitive load assessment in industrial tasks. Robotics Computer-
time. Appl. Ergon. 32, 357–366. doi:10.1016/s0003-6870(01)00006-0
Integrated Manuf. 78, 102380. doi:10.1016/j.rcim.2022.102380
Kemmlert, K. (1995). A method assigned for the identification of ergonomic
Lagomarsino, M., Lorenzini, M., De Momi, E., and Ajoudani, A. (2022c). “Robot
hazards—Plibel. Appl. Ergon. 26, 199–211. doi:10.1016/0003-6870(95)00022-5
trajectory adaptation to optimise the trade-off between human cognitive
Keyserling, W., Brouwer, M., and Silverstein, B. (1992). A checklist for evaluating ergonomics and workplace productivity in collaborative tasks,” in Proceedings
ergonomic risk factors resulting from awkward postures of the legs, trunk and neck. of International Conference on Intelligent Robots and Systems (IROS), Kyoto,
Int. J. Industrial Ergonomics 9, 283–301. doi:10.1016/0169-8141(92)90062-5 Japan (IEEE).
Kim, W., Garate, V. R., Gandarias, J. M., Lorenzini, M., and Ajoudani, A. (2021a). Lambrechts, W., Klaver, J. S., Koudijzer, L., and Semeijn, J. (2021). Human factors
A directional vibrotactile feedback interface for ergonomic postural adjustment. influencing the implementation of cobots in high volume distribution centres.
IEEE Trans. Haptics 1, 200–211. doi:10.1109/toh.2021.3112795 Logistics 5, 32. doi:10.3390/logistics5020032
Kim, W., Lee, J., Peternel, L., Tsagarakis, N., and Ajoudani, A. (2018a). Lamon, E., De Franco, A., Peternel, L., and Ajoudani, A. (2019). A capability-
Anticipatory robot assistance for the prevention of human static joint aware role allocation approach to industrial assembly tasks. IEEE Robot. Autom.
overloading in human–robot collaboration. IEEE Robot. Autom. Lett. 3, 68–75. Lett. 4, 3378–3385. doi:10.1109/lra.2019.2926963
doi:10.1109/lra.2017.2729666
Landi, C. T., Villani, V., Ferraguti, F., Sabattini, L., Secchi, C., and Fantuzzi, C.
Kim, W., Lorenzini, M., Balatti, P., Nguyen, P. D., Pattacini, U., Tikhanoff, V., (2018). Relieving operators’ workload: Towards affective robotics in industrial
et al. (2019). Adaptable workstations for human-robot collaboration: A scenarios. Mechatronics 54, 144–154. doi:10.1016/j.mechatronics.2018.07.012
reconfigurable framework for improving worker ergonomics and productivity.
Latella, C., Traversaro, S., Ferigo, D., Tirupachuri, Y., Rapetti, L., Andrade
IEEE Robot. Autom. Mag. 26, 14–26. doi:10.1109/mra.2018.2890460
Chavez, F. J., et al. (2019). Simultaneous floating-base estimation of human
Kim, W., Lorenzini, M., Kapıcıoğlu, K., and Ajoudani, A. (2018b). Ergotac: A kinematics and joint torques. Sensors 19, 2794. doi:10.3390/s19122794
tactile feedback interface for improving human ergonomics in workplaces. IEEE
Laudante, E., Greco, A., Caterino, M., and Fera, M. (2020). Human–robot
Robot. Autom. Lett. 3, 4179–4186. doi:10.1109/lra.2018.2864356
interaction for improving fuselage assembly tasks: A case study. Appl. Sci. 10,
Kim, W., Peternel, L., Lorenzini, M., Babič, J., and Ajoudani, A. (2021b). A 5757. doi:10.3390/app10175757
human-robot collaboration framework for improving ergonomics during dexterous
Leppink, J., Paas, F., der Vleuten, C. P. M. V., Gog, T. V., and Merriënboer, J. J. G.
operation of power tools. Robotics Computer-Integrated Manuf. 68, 102084. doi:10.
V. (2013). Development of an instrument for measuring different types of cognitive
1016/j.rcim.2020.102084
load. Behav. Res. Methods 45, 1058–1072. doi:10.3758/s13428-013-0334-1
Kim, W., Sung, J., Saakes, D., Huang, C., and Xiong, S. (2021c). Ergonomic
Li, G., and Buckle, P. (1998). A practical method for the assessment of work-
postural assessment using a new open-source human pose estimation technology
related musculoskeletal risks - quick exposure check (QEC). Proc. Hum.
(OpenPose). Int. J. Industrial Ergonomics 84, 103164. doi:10.1016/j.ergon.2021.
Factors Ergonomics Soc. Annu. Meet. 42, 1351–1355. doi:10.1177/
103164
154193129804201905
Kleinsmith, A., and Bianchi-Berthouze, N. (2013). Affective body expression
Li, G., and Buckle, P. (1999). Current techniques for assessing physical exposure
perception and recognition: A survey. IEEE Trans. Affect. Comput. 4, 15–33. doi:10.
to work-related musculoskeletal risks, with emphasis on posture-based methods.
1109/t-affc.2012.16
Ergonomics 42, 674–695. doi:10.1080/001401399185388
Klussmann, A., Liebers, F., Gebhardt, H., Rieger, M. A., Latza, U., and Steinberg,
Liau, Y. Y., and Ryu, K. (2020). Task allocation in human-robot collaboration
U. (2017). Risk assessment of manual handling operations at work with the key
(hrc) based on task characteristics and agent capability for mold assembly. Procedia
indicator method (kim-mho)—Determination of criterion validity regarding the
Manuf. 51, 179–186. doi:10.1016/j.promfg.2020.10.026
prevalence of musculoskeletal symptoms and clinical conditions within a cross-
sectional study. BMC Musculoskelet. Disord. 18 (1), 1–13. doi:10.1186/s12891-017- Lietaert, P., Billen, N., and Burggraeve, S. (2019). “Model-based multi-attribute
1542-0 collaborative production cell layout optimization,” in Proceedings of International
Conference on Research and Education in Mechatronics (REM), Wels, Austria
Kolus, A., Wells, R., and Neumann, P. (2018). Production quality and human
(IEEE), 1–7.
factors engineering: A systematic review and theoretical framework. Appl. Ergon.
73, 55–89. doi:10.1016/j.apergo.2018.05.010 Liu, H., and Wang, L. (2020). Remote human–robot collaboration: A
cyber–physical system application for hazard manufacturing environment.
Krupke, D., Steinicke, F., Lubos, P., Jonetzko, Y., Görner, M., and Zhang, J. (2018).
J. Manuf. Syst. 54, 24–34. doi:10.1016/j.jmsy.2019.11.001
“Comparison of multimodal heading and pointing gestures for co-located mixed

Lloyd, D. G., and Besier, T. F. (2003). An emg-driven musculoskeletal model to Michalos, G., Spiliotopoulos, J., Makris, S., and Chryssolouris, G. (2018). A
estimate muscle forces and knee joint moments in vivo. J. biomechanics 36, 765–776. method for planning human robot shared tasks. CIRP J. Manuf. Sci. Technol. 22,
doi:10.1016/s0021-9290(03)00010-1 76–90. doi:10.1016/j.cirpj.2018.05.003
Lopez-Nava, I. H., and Angelica, M. M. (2016). Wearable inertial sensors for Mientjes, M. I., Norman, R., Wells, R., and McGill, S. (1999). Assessment of an
human motion analysis: A review. IEEE Sens. J. 16, 7821–7834. doi:10.1109/jsen. emg-based method for continuous estimates of low back compression during
2016.2609392 asymmetrical occupational tasks. Ergonomics 42, 868–879. doi:10.1080/
001401399185342
Lorenzini, M., Kim, W., and Ajoudani, A. (2022). An online multi-index
approach to human ergonomics assessment in the workplace. IEEE Trans. Mijović, P., Ković, V., De Vos, M., Mačužić, I., Todorović, P., Jeremić, B., et al.
Hum. Mach. Syst., 1–12. doi:10.1109/thms.2021.3133807 (2017). Towards continuous and real-time attention monitoring at work: Reaction
time versus brain response. Ergonomics 60, 241–254. doi:10.1080/00140139.2016.
Lorenzini, M., Kim, W., De Momi, E., and Ajoudani, A. (2019). “A new
1142121
overloading fatigue model for ergonomic risk assessment with application to
human-robot collaboration,” in Proceedings of International Conference on Millard, M., Uchida, T., Seth, A., and Delp, S. (2013). Flexing computational
Robotics and Automation (ICRA), Montreal, Canada (IEEE), 1962–1968. muscle: Modeling and simulation of musculotendon dynamics. J. Biomech. Eng.
135, 021005. doi:10.1115/1.4023390
Lorenzini, M., Kim, W., De Momi, E., and Ajoudani, A. (2018). A synergistic
approach to the real-time estimation of the feet ground reaction forces and centers Monnington, S., Quarrie, C., Pinder, A., and Morris, L. (2003). Development of
of pressure in humans with application to human–robot collaboration. IEEE Robot. manual handling assessment charts (MAC) for health and safety inspectors.
Autom. Lett. 3, 3654–3661. doi:10.1109/lra.2018.2855802 Contemp. Ergon., 3–8.
Ma, L., Chablat, D., Bennis, F., and Zhang, W. (2009). A new simple dynamic Mündermann, L., Corazza, S., and Andriacchi, T. P. (2006). The evolution of methods
muscle fatigue model and its validation. Int. J. Industrial Ergonomics 39, 211–220. for the capture of human movement leading to markerless motion capture for
doi:10.1016/j.ergon.2008.04.004 biomechanical applications. J. Neuroeng. Rehabil. 3, 6. doi:10.1186/1743-0003-3-6
Mainprice, J., Gharbi, M., Siméon, T., and Alami, R. (2012). “Sharing effort in Murai, A., Kurosaki, K., Yamane, K., and Nakamura, Y. (2010). Musculoskeletal-
planning human-robot handover tasks,” in Proceedings of International see-through mirror: Computational modeling and algorithm for whole-body
Conference on Robot and Human Interactive Communication (RO-MAN), muscle activity visualization in real time. Prog. biophysics Mol. Biol. 103,
Paris, France (IEEE), 764–770. 310–317. doi:10.1016/j.pbiomolbio.2010.09.006
Mainprice, J., Sisbot, E. A., Jaillet, L., Cortés, J., Alami, R., and Siméon, T. (2011). Muzaffar, S., and Elfadel, I. A. M. (2020). Shoe-integrated, force sensor design for
“Planning human-aware motions using a sampling-based costmap planner,” in continuous body weight monitoring. Sensors 20, 3339. doi:10.3390/s20123339
Proceedings of International Conference on Robotics and Automation (ICRA),
Nagymáté, G., and Kiss, R. M. (2018). Application of OptiTrack motion capture
Shanghai, China (IEEE), 5012–5017.
systems in human movement analysis. RIiM. 5, 1–9. doi:10.17667/riim.2018.1/13
Malaise, A., Maurice, P., Colas, F., and Ivaldi, S. (2019). Activity recognition for
Naismith, L. M., Cheung, J. J. H., Ringsted, C., and Cavalcanti, R. B. (2015).
ergonomics assessment of industrial tasks with automatic feature selection. IEEE
Limitations of subjective cognitive load measures in simulation-based procedural
training. Med. Educ. 49, 805–814. doi:10.1111/medu.12732
Manal, K., Gonzalez, R. V., Lloyd, D. G., and Buchanan, T. S. (2002). A real-time
Nakamura, Y., Yamane, K., Fujita, Y., and Suzuki, I. (2005). Somatosensory
emg-driven virtual arm. Comput. Biol. Med. 32, 25–36. doi:10.1016/s0010-4825(01)
computation for man-machine interface from motion-capture data and
00024-5
musculoskeletal human model. IEEE Trans. Robot. 21, 58–66. doi:10.1109/tro.
Manal, K., Gravare-Silbernagel, K., and Buchanan, T. S. (2012). A real-time emg- 2004.833798
driven musculoskeletal model of the ankle. Multibody Syst. Dyn. 28, 169–180.
Nicora, M. L., André, E., Berkmans, D., Carissoli, C., D’Orazio, T., Fave, A. D., et al.
doi:10.1007/s11044-011-9285-4
(2021). A human-driven control architecture for promoting good mental health in
Marin, A. G., Shourijeh, M. S., Galibarov, P. E., Damsgaard, M., Fritzsch, L., and collaborative robot scenarios. IEEE, 285–291. doi:10.1109/RO-MAN50785.2021.9515315
Stulp, F. (2018). “Optimizing contextual ergonomics models in human-robot
Niemann, F., Reining, C., Rueda, F. M., Nair, N. R., Steffens, J. A., Fink, G. A., et al.
interaction,” in Proceedings of International Conference on Intelligent Robots
(2020). LARa: Creating a dataset for human activity recognition in logistics using
and Systems (IROS, Madrid, Spain (IEEE), 1–9.
semantic attributes. Sensors 20, 4083. doi:10.3390/s20154083
Marín, J., and Marín, J. J. (2021). Forces: A motion capture-based ergonomic
Nikooyan, A., Veeger, H., Westerhoff, P., Bolsterlee, B., Graichen, F., Bergmann,
method for the today’s world. Sensors 21, 5139. doi:10.3390/s21155139
G., et al. (2012). An emg-driven musculoskeletal model of the shoulder. Hum. Mov.
Marras, W. S. (2006). Fundamentals and assessment tools for occupational Sci. 31, 429–447. doi:10.1016/j.humov.2011.08.006
ergonomics. Boca Raton, FL: CRC Press.
Novak, D., Mihelj, M., and Munih, M. (2011). Psychophysiological responses to
Mateus, J. C., Claeys, D., Limère, V., Cottyn, J., and Aghezzaf, E.-H. (2019). A structured different levels of cognitive and physical workload in haptic interaction. Robotica
methodology for the design of a human-robot collaborative assembly workplace. Int. 29, 367–374. doi:10.1017/s0263574710000184
J. Adv. Manuf. Technol. 102, 2663–2681. doi:10.1007/s00170-019-03356-3
Occhipinti, E. (1998). Ocra: A concise index for the assessment of exposure to
Maurice, P., Malaisé, A., Amiot, C., Paris, N., Richard, G.-J., Rochel, O., et al. repetitive movements of the upper limbs. Ergonomics 41, 1290–1311. doi:10.1080/
(2019). Human movement and ergonomics: An industry-oriented dataset for 001401398186315
collaborative robotics. Int. J. Rob. Res. 38, 1529–1537. doi:10.1177/
Paas, F., Tuovinen, J. E., Tabbers, H., and Gerven, P. W. M. V. (2003). Cognitive
0278364919882089
load measurement as a means to advance cognitive load theory. Educ. Psychol. 38,
Maurice, P., Padois, V., Measson, Y., and Bidaud, P. (2017). Human-oriented 63–71. doi:10.1207/s15326985ep3801_8
design of collaborative robots. Int. J. Industrial Ergonomics 57, 88–102. doi:10.1016/
Palomba, I., Gualtieri, L., Rojas, R., Rauch, E., Vidoni, R., and Ghedin, A. (2021).
j.ergon.2016.11.011
Mechatronic re-design of a manual assembly workstation into a collaborative one
McAtamney, L., and Corlett, E. N. (1993). Rula: A survey method for the for wire harness assemblies. Robotics 10, 43. doi:10.3390/robotics10010043
investigation of work-related upper limb disorders. Appl. Ergon. 24, 91–99.
Parastegari, S., Abbasi, B., Noohi, E., and Zefran, M. (2017). “Modeling human
doi:10.1016/0003-6870(93)90080-s
reaching phase in human-human object handover with application in robot-human
Menolotto, M., Komaris, D.-S., Tedesco, S., O’Flynn, B., and Walsh, M. (2020). handover,” in Proceedings of International Conference on Intelligent Robots and
Motion capture technology in industrial applications: A systematic review. Sensors Systems (IROS), Vancouver, Canada (IEEE), 3597–3602.
20, 5687. doi:10.3390/s20195687
Park, J., Heo, P., Kim, J., and Na, Y. (2019). A finger grip force sensor with an
Merlo, E., Lamon, E., Fusaro, F., Lorenzini, M., Carfì, A., Mastrogiovanni, F., et al. open-pad structure for glove-type assistive devices. Sensors 20, 4. doi:10.3390/
(2021). Dynamic human-robot role allocation based on human ergonomics risk s20010004
prediction and robot actions adaptation. arXiv preprint arXiv:2111.03630.
Pascual, S. A., and Naqvi, S. (2008). An investigation of ergonomics analysis tools
Messeri, C., Masotti, G., Zanchettin, A. M., and Rocco, P. (2021). Human-robot used in industry in the identification of work-related musculoskeletal disorders. Int.
collaboration: Optimizing stress and productivity based on game theory. IEEE J. Occup. Saf. Ergonomics 14, 237–245. doi:10.1080/10803548.2008.11076755
Patrizi, A., Pennestrì, E., and Valentini, P. P. (2016). Comparison between low-
Meyer, A. J., Patten, C., and Fregly, B. J. (2017). Lower extremity emg-driven cost marker-less and high-end marker-based motion capture systems for the
modeling of walking with automated adjustment of musculoskeletal geometry. PloS computer-aided assessment of working ergonomics. Ergonomics 59, 155–162.
one 12, e0179698. doi:10.1371/journal.pone.0179698 doi:10.1080/00140139.2015.1057238

Pau, J. W., Xie, S. S., and Pullan, A. J. (2012). Neuromuscular interfacing: Schaub, K., Caragnano, G., Britzke, B., and Bruder, R. (2013). The European
Establishing an emg-driven model for the human elbow joint. IEEE Trans. assembly worksheet. Theor. Issues Ergonomics Sci. 14, 616–639. doi:10.1080/
Biomed. Eng. 59, 2586–2593. doi:10.1109/tbme.2012.2206389 1463922x.2012.678283
Pearce, M., Mutlu, B., Shah, J., and Radwin, R. (2018). Optimizing makespan and Setz, C., Arnrich, B., Schumm, J., Marca, R. L., Tröster, G., and Ehlert, U. (2010).
ergonomics in integrating collaborative robots into manufacturing processes. IEEE Discriminating stress from cognitive load using a wearable EDA device. IEEE Trans.
Trans. Autom. Sci. Eng. 15, 1772–1784. doi:10.1109/tase.2018.2789820 Inf. Technol. Biomed. 14, 410–417. doi:10.1109/titb.2009.2036164
Peruzzini, M., Pellicciari, M., and Gadaleta, M. (2019). A comparative study on Sgarbossa, F., Grosse, E. H., Neumann, W. P., Battini, D., and Glock, C. H. (2020).
computer-integrated set-ups to design human-centred manufacturing systems. Human factors in production and logistics systems of the future. Annu. Rev. Control
Robotics Computer-Integrated Manuf. 55, 265–278. doi:10.1016/j.rcim.2018.03.009 49, 295–305. doi:10.1016/j.arcontrol.2020.04.007
Peternel, L., Fang, C., Tsagarakis, N., and Ajoudani, A. (2019). A selective muscle Shackel, B., Chidsey, K., and Shipley, P. (1969). The assessment of chair comfort.
fatigue management approach to ergonomic human-robot co-manipulation. Ergonomics 12, 269–306. doi:10.1080/00140136908931053
Robotics Computer-Integrated Manuf. 58, 69–79. doi:10.1016/j.rcim.2019.01.013
Shafti, A., Ataka, A., Lazpita, B. U., Shiva, A., Wurdemann, H. A., and Althoefer,
Peternel, L., Tsagarakis, N., and Ajoudani, A. (2017). A human–robot co- K. (2019). “Real-time robot-assisted ergonomics,” in Proceedings of International
manipulation approach based on human sensorimotor information. IEEE Trans. Conference on Robotics and Automation (ICRA), Montreal, Canada (IEEE),
Neural Syst. Rehabil. Eng. 25, 811–822. doi:10.1109/tnsre.2017.2694553 1975–1981.
Peternel, L., Tsagarakis, N., Caldwell, D., and Ajoudani, A. (2018). Robot Shaikh, I., Kim, Y., Jayaram, S., Jayaram, U., and Choi, H. (2003). “Integration
adaptation to human physical fatigue in human–robot co-manipulation. Auton. of immersive environment and rula for real-time study of workplace related
Robots 42, 1011–1021. doi:10.1007/s10514-017-9678-1 musculoskeletal disorders in the upper limb,” in Proceedings of
International Design Engineering Technical Conferences and
Plantard, P., Shum, H., Le Pierres, A., and Multon, F. (2017). Validation of an
Computers and Information in Engineering Conference, Chicago, Illinois,
ergonomic assessment method using kinect data in real workplace conditions. Appl.
USA, 1163–1171.
Ergon. 65, 562–569. doi:10.1016/j.apergo.2016.10.015
Shikdar, A. A., and Hadhrami, M. A. A. (2007). Smart workstation design: An
Poh, M.-Z., Swenson, N. C., and Picard, R. W. (2010). A wearable sensor for
ergonomics and methods engineering approach. Int. J. Industrial Syst. Eng. 2, 363.
unobtrusive, long-term assessment of electrodermal activity. IEEE Trans. Biomed.
doi:10.1504/ijise.2007.013184
Eng. 57, 1243–1252. doi:10.1109/tbme.2009.2038487
Sisbot, E. A., and Alami, R. (2012). A human-aware manipulation planner. IEEE
Priel, V. Z. (1974). A numerical definition of posture. Hum. Factors 16, 576–584.
Trans. Robot. 28, 1045–1057. doi:10.1109/tro.2012.2196303
doi:10.1177/001872087401600602
Snook, S., and Ciriello, V. (1991). The design of manual handling tasks: Revised
Puthenveetil, S., Daphalapurkar, C., Zhu, W., Leu, M., Liu, X., Gilpin-Mcminn, J.,
tables of maximum acceptable weights and forces. Ergonomics 34, 1197–1213.
et al. (2015). Computer-automated ergonomic analysis based on motion capture
doi:10.1080/00140139108964855
and assembly simulation. Virtual Real. 19, 119–128. doi:10.1007/s10055-015-
0261-9 So, W. K. Y., Wong, S. W. H., Mak, J. N., and Chan, R. H. M. (2017). An
evaluation of mental workload with frontal EEG. PLOS ONE 12, e0174949. doi:10.
Rajavenkatanarayanan, A., Nambiappan, H. R., Kyrarini, M., and Makedon, F.
1371/journal.pone.0174949
(2020). “Towards a real-time cognitive load assessment system for industrial
human-robot cooperation,” in Proceedings of International Conference on Sony, M., and Naik, S. (2019). Critical factors for the successful implementation of
Robot and Human Interactive Communication (RO-MAN), Naples Italy (IEEE), industry 4.0: A review and future research direction. Prod. Plan. Control 31,
698–705. 799–815. doi:10.1080/09537287.2019.1691278
Ranavolo, A., Draicchio, F., Varrecchia, T., Silvetti, A., and Iavicoli, S. (2018). Sony, M., and Naik, S. (2020). Industry 4.0 integration with socio-technical
Wearable monitoring devices for biomechanical risk assessment at work: Current systems theory: A systematic review and proposed theoretical model. Technol. Soc.
status and future challenges—A systematic review. Int. J. Environ. Res. Public Health 61, 101248. doi:10.1016/j.techsoc.2020.101248
15, 2001. doi:10.3390/ijerph15092001
Steinberg, U. (2012). New tools in Germany: Development and appliance of the
Rapetti, L., Tirupachuri, Y., Nava, G., Latella, C., Darvish, K., and Pucci, D. first two kim (” lifting, holding and carrying” and” pulling and pushing”) and
(2019). “Partner-aware humanoid robot control: From robot-robot collaboration to practical use of these methods. Work 41, 3990–3996. doi:10.3233/wor-2012-0698-
human-robot collaboration and ergonomy control,” in Proceedings of International 3990
Conference on Intelligent Robots and Systems (IROS), Macau, China (New York,
Steven Moore, J., and Garg, A. (1995). The strain index: A proposed method to
NY: IEEE). Workshop on Progress in Ergonomic Physical Human-Robot
analyze jobs for risk of distal upper extremity disorders. Am. Industrial Hyg. Assoc. J.
Interaction.
56, 443–458. doi:10.1080/15428119591016863
Ray, S., and Teizer, J. (2012). Real-time construction worker posture analysis for
Stroeve, S. (1999). Impedance characteristics of a neuromusculoskeletal model of
ergonomics training. Adv. Eng. Inf. 26, 439–455. doi:10.1016/j.aei.2012.02.011
the human arm i. posture control. Biol. Cybern. 81, 475–494. doi:10.1007/
Reid, G. B., and Nygren, T. E. (1988). The subjective workload assessment s004220050577
technique: A scaling procedure for measuring mental workload. Adv. Psychol.
Sun, X., Houssin, R., Renaud, J., and Gardoni, M. (2019). A review of
52, 185–218. doi:10.1016/s0166-4115(08)62387-0
methodologies for integrating human factors and ergonomics in engineering
Rodgers, S. H. (2004). “Muscle fatigue assessment: Functional job analysis design. Int. J. Prod. Res. 57, 4961–4976. doi:10.1080/00207543.2018.1492161
technique,” in Handbook of human factors and ergonomics methods (Boca
Sweller, J., van Merrienboer, J. J. G., and Paas, F. G. W. C. (1998). Cognitive
Raton, FL: CRC Press), 130–141.
architecture and instructional design. Educ. Psychol. Rev. 10, 251–296. doi:10.1023/
Rojas, R. A., Garcia, M. A. R., Gualtieri, L., Wehrle, E., Rauch, E., and Vidoni, R. a:1022193728205
(2021). “Automatic planning of psychologically less-stressful trajectories in
Szabò, G., and Dobò, M. (2018). CERA, an integrated tool for ergonomic risk
collaborative workstations: An integrated toolbox for unskilled users,” in
assessment. dtssehs. doi:10.12783/dtssehs/ise2018/33661
Proceedings of Symposium on Robot Design, Dynamics and Control, Sapporo,
Japan (Springer), 118–126. Thompson, E. R. (2007). Development and validation of an internationally
reliable short-form of the positive and negative affect schedule (panas). J. cross-
Rojas, R. A., Garcia, M. A. R., Wehrle, E., and Vidoni, R. (2019). A variational
cultural Psychol. 38, 227–242. doi:10.1177/0022022106297301
approach to minimum-jerk trajectories for psychological safety in collaborative
assembly stations. IEEE Robot. Autom. Lett. 4, 823–829. doi:10.1109/lra.2019. Thorvald, P., Lindblom, J., and Andreasson, R. (2019). On the
2893018 development of a method for cognitive load assessment in manufacturing.
Robotics Computer-Integrated Manuf. 59, 252–266. doi:10.1016/j.rcim.2019.
Roy, T., Marwala, T., and Chakraverty, S. (2020). Advancements and role of
04.012
emotion recognition in the 4th industrial revolution. Disruptive Fourth Industrial
Revolut. Lect. Notes Electr. Eng. 674, 179–203. Van de Perre, G., El Makrini, I., Van Acker, B., Saldien, J., Vergara, C., Pintelon,
L., et al. (2018). “Improving productivity and worker conditions in assembly part 1:
Rubio, S., Diaz, E., Martin, J., and Puente, J. M. (2004). Evaluation of subjective
A collaborative architecture and task allocation framework,” in IROS Workshop on
mental workload: A comparison of SWAT, NASA-TLX, and workload profile
Robotic Co-workers 4.0: Human Safety and Comfort in Human-Robot Interactive
methods. Appl. Psychol. 53, 61–86. doi:10.1111/j.1464-0597.2004.00161.x
Social Environments, Madrid, Spain (IEEE).
Sartori, M., Reggiani, M., Farina, D., and Lloyd, D. G. (2012). Emg-driven
van der Beek, A. J., and Frings-Dresen, M. (1998). Assessment of mechanical
forward-dynamic estimation of muscle force and joint moment about multiple
exposure in ergonomic epidemiology. Occup. Environ. Med. 55, 291–299. doi:10.
degrees of freedom in the human lower extremity. PloS one 7, e52618. doi:10.1371/
1136/oem.55.5.291
journal.pone.0052618

van der Spaa, L., Gienger, M., Bates, T., and Kober, J. (2020). “Predicting and Winkel, J., and Mathiassen, S. E. (1994). Assessment of physical work load in
optimizing ergonomics in physical human-robot cooperation tasks,” in Proceedings epidemiologic studies: Concepts, issues and operational considerations. Ergonomics
of International Conference on Robotics and Automation (ICRA), Paris, France (IEEE), 37, 979–988. doi:10.1080/00140139408963711
1799–1805.
Winter, D. A. (2009). Biomechanics and motor control of human movement. John
Ventura, L., Lorenzini, M., Kim, W., and Ajoudani, A. (2021). A flexible robotics- Wiley & Sons. chap. 4: Anthropometry.
inspired computational model of compressive loading on the human spine. IEEE
Wu, X., and Li, Z. (2013). “Secondary task method for workload measurement
in alarm monitoring and identification tasks,” in Cross-Cultural
Vidullch, M. A., Ward, G. F., and Schueren, J. (1991). Using the subjective Design. Methods, Practice, and Case Studies, Berlin, Germany (Springer),
workload dominance (SWORD) technique for projective workload assessment. 346–354.
Hum. Factors 33, 677–691. doi:10.1177/001872089103300605
Wurhofer, D., Meneweger, T., Fuchsberger, V., and Tscheligi, M. (2015).
Vignais, N., Miezal, M., Bleser, G., Mura, K., Gorecky, D., and Marin, F. (2013). “Deploying robots in a production environment: A study on temporal
Innovative system for real-time ergonomic feedback in industrial manufacturing. transitions of workers’ experiences,” in Proceedings of Human-Computer
Appl. Ergon. 44, 566–574. doi:10.1016/j.apergo.2012.11.008 Interaction – INTERACT, Bamberg, Germany (Springer International
Publishing), 203–220.
Village, J., Frazer, M., Cohen, M., Leyland, A., Park, I., and Yassi, A. (2005).
Electromyography as a measure of peak and cumulative workload in Xie, B., and Salvendy, G. (2000). Prediction of mental workload in single and
intermediate care and its relationship to musculoskeletal injury: An multiple tasks environments. Int. J. Cognitive Ergonomics 4, 213–242. doi:10.1207/
exploratory ergonomic study. Appl. Ergon. 36, 609–618. doi:10.1016/j.apergo. s15327566ijce0403_3
2005.01.019
Yahya, M., Shah, J. A., Kadir, K. A., Yusof, Z. M., Khan, S., and Warsi, A.
Villani, V., Capelli, B., Secchi, C., Fantuzzi, C., and Sabattini, L. (2020). Humans (2019). Motion capture sensing techniques used in human upper limb motion:
interacting with multi-robot systems: A natural affect-based approach. Auton. A review. Sens. Rev. 39, 504–511. doi:10.1108/sr-10-2018-0270
Robots 44, 601–616. doi:10.1007/s10514-019-09889-6
Yang, S., Yin, Z., Wang, Y., Zhang, W., Wang, Y., and Zhang, J. (2019). Assessing
Villani, V., Pini, F., Leali, F., and Secchi, C. (2018a). Survey on human–robot cognitive mental workload via EEG signals and an ensemble deep learning classifier
collaboration in industrial settings: Safety, intuitive interfaces and based on denoising autoencoders. Comput. Biol. Med. 109, 159–170. doi:10.1016/j.
applications. Mechatronics 55, 248–266. doi:10.1016/j.mechatronics.2018. compbiomed.2019.04.034
02.009
Yin, B., Chen, F., Ruiz, N., and Ambikairajah, E. (2008). “Speech-based
Villani, V., Sabattini, L., Secchi, C., and Fantuzzi, C. (2018b). “A framework for cognitive load monitoring system,” in Proceedings of International Conference
affect-based natural human-robot interaction,” in Proceedings of International on Acoustics, Speech and Signal Processing (ICASSP), Las Vegas, NV, USA
Conference on Robot and Human Interactive Communication (RO-MAN), (IEEE), 2041–2044.
Nanjing and Tai’an, China: (IEEE), 10.
Zacharaki, A., Kostavelis, I., Gasteratos, A., and Dokas, I. (2020). Safety bounds in
Waters, T. R., Putz-Anderson, V., Garg, A., and Fine, L. J. (1993). Revised niosh human robot interaction: A survey. Saf. Sci. 127, 104667. doi:10.1016/j.ssci.2020.104667
equation for the design and evaluation of manual lifting tasks. Ergonomics 36,
Zacharias, F., Schlette, C., Schmidt, F., Borst, C., Rossmann, J., and Hirzinger, G.
749–776. doi:10.1080/00140139308967940
(2011). “Making planned paths look more human-like in humanoid robot
Westgaard, R., and Winkel, J. (1996). Guidelines for occupational manipulation planning,” in Proceedings of International Conference on Robotics
musculoskeletal load as a basis for intervention: A critical review. Appl. Ergon. and Automation (ICRA), Shanghai, China (IEEE), 1192–1198.
27, 79–88. doi:10.1016/0003-6870(95)00062-3
Zanchettin, A. M., Lotano, E., and Rocco, P. (2019). “Collaborative
Wiktorin, C., Karlqvist, L., Winkel, J., and Group, S. M. I. S. (1993). Validity of self- robot assistant for the ergonomic manipulation of cumbersome objects,”
reported exposures to work postures and manual materials handling. Scand. J. work, in Proceedings of International Conference
208–214. on Intelligent Robots and Systems (IROS), Macau, China (IEEE), 6729–6734.

Frobt 09 813907

Uploaded by

Copyright:

Available Formats

Frobt 09 813907

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Frobt 09 813907

Uploaded by

Copyright:

Available Formats

TYPE Review

PUBLISHED 03 January 2023

Frontiers in Robotics and AI 01 frontiersin.org

Frontiers in Robotics and AI 02 frontiersin.org

Frontiers in Robotics and AI 03 frontiersin.org

Subjective measurements are surveys that can be completed

Frontiers in Robotics and AI 04 frontiersin.org

Frontiers in Robotics and AI 05 frontiersin.org

Frontiers in Robotics and AI 06 frontiersin.org

Frontiers in Robotics and AI 07 frontiersin.org

Frontiers in Robotics and AI 08 frontiersin.org

It can be noted that these indicators have commonalities with

Frontiers in Robotics and AI 09 frontiersin.org

Frontiers in Robotics and AI 10 frontiersin.org

Frontiers in Robotics and AI 11 frontiersin.org

Frontiers in Robotics and AI 12 frontiersin.org

Frontiers in Robotics and AI 13 frontiersin.org

Frontiers in Robotics and AI 14 frontiersin.org

Physical ergonomics in Systematic observations Direct measurements

Posture-based Comprehensive Motion and forces Electromyography

Physiological Traditional Peternel et al. (2017), Peternel

Frontiers in Robotics and AI 15 frontiersin.org

Cognitive ergonomics in Systematic observations Direct measurements

Frontiers in Robotics and AI 16 frontiersin.org

Frontiers in Robotics and AI 17 frontiersin.org

interface technologies. These technologies, which are primarily Author contributions

Frontiers in Robotics and AI 18 frontiersin.org

Frontiers in Robotics and AI 19 frontiersin.org

Frontiers in Robotics and AI 20 frontiersin.org

Frontiers in Robotics and AI 21 frontiersin.org

Frontiers in Robotics and AI 22 frontiersin.org

Frontiers in Robotics and AI 23 frontiersin.org

Frontiers in Robotics and AI 24 frontiersin.org

You might also like