IEEE Robotics & Automation Magazine - March 2020 - 55

activations of the upper-limb muscles by up to 43%. Significant reductions were observed for most upper-limb muscles
involved in flexion movements and shoulder stabilization,
particularly the anterior deltoid, medial deltoid, trapezius
ascendens, and pectoralis major. The shoulder's ranges of
motion for abduction-adduction and flexion-extension were
not significantly altered when using the proto-MATE, and the
human-robot relative displacement during the tasks was
always lower than 2 mm, thus proving a stable and reliable
human-robot kinematics coupling. Further longitudinal
studies with workers are needed to investigate the long-term
effects of using the device on the incidence of work-related
musculoskeletal disorders (WRMDs).
Background
According to the Sixth European Working Conditions Survey, WRMDs are the most common occupational illnesses
in Europe. Indeed, nearly 50% of European workers (75-
80 million) suffer from back, neck, or upper-limb disorders [1], causing significant health and cost issues [2].
These health issues cause a huge financial burden for
companies and healthcare systems. Indeed, companies
must manage the costs of replacing and training workers as
well as the related reduced productivity, while healthcare
systems need to address the costs for all compensation
claims made [2].
To reduce exposure to the risk factors of musculoskeletal
injuries, three main courses of actions can be considered: 1)
redesigning the workplace (e.g., optimizing the labor environment to enable working in a comfortable, strain-free posture); 2) integrating job rotations; and 3) introducing
practices such as training for workers or periodic rest periods. Despite the approaches mentioned, repetitive movements, awkward postures, overexertion, and vibration can
still cause a significant number of WRMDs. Most disorders
on record are related to repetitive movements (which
account for 61% of the workers); of these, 44% are related to
overhead tasks [3] and awkward body postures (which
account for 43% of WRMDs cases) [1].
Substituting humans with robots in repetitive tasks related
to upper-limb activities would seem like a possible solution to
safeguard humans from performing strenuous and repetitive
movements while maintaining productivity. However, many
specific tasks involving shoulder elevation require high flexibility and versatility, and fully automated solutions are not yet
feasible. In the last few years, several companies started to
develop and commercialize gravity-assistive, wearable upperlimb exoskeletons to improve conditions for workers and follow the recommendations provided by European Union
legislations in terms of assessing and reducing workers' ergonomics risks (e.g., UNI EN 1005-4 related to incongruent
postures and UNI EN 1005-pr5 related to repetitive handling
at high frequency).
Given the strict health and safety requirements with
which any industrial equipment must comply and considering several psychological and physical barriers, most

companies opted to develop passive devices. Examples of
commercially available upper-limb passive exoskeletons for
workers are EksoVest (EksoBionics, Richmond, California),
Airframe (Levitate Technologies, San Diego, California),
ShoulderX (SuitX Emeryville, California), PAEXO (Ottobock, Duderstadt, Germany), and Skel'Ex (Skel'Ex, Rotterdam, The Netherlands).
Typically, the effects of upper-limb exoskeletons for worker applications have been evaluated in tests where users perform specific tasks that simulate potential applications (i.e.,
light assembly, overhead drilling, wiring, or painting). In these
scenarios, the metrics most often used to verify the device's
effectiveness in reducing
users' physical demands
relate to the reduced activThe MATE was designed
ity of the assisted muscles
[4]-[7]. In some cases,
to provide a compact
improvements in the precision and quantity of
design thus minimizing
work tasks have been
reported as effectivenessthe risk of entrapment and
related metrics [8]. In
addition to effectiveness
entanglement with parts of
metrics, Theurel et al. [7]
and Kim et al. [5] investithe external environment.
gated potential alterations
of the arm kinematics or
body posture while standing or walking with the device as a way to verify potential
undesired effects of the exoskeletons. With the same objective, Van Engelhoven et al. [6] and Theurel et al. [7] measured
the biomechanical strain on unassisted muscles. Several studies also reported on subjective evaluations of the device, by
means of acceptance questionnaires [e.g., the Technology
Acceptance Model (TAM) and TAM2 [8]] and usability questionnaires (the System Usability Scale). In some studies, the
perceived discomfort of selected body parts and the overall
perceived workload when executing the task (NASA Task
Load Index) were also considered [5].
We present and evaluate a novel passive exoskeleton
designed to provide upper-limb support to workers in overhead tasks and repetitive upper-limb movements. The device
is characterized by two distinguishing design features: a highly ergonomic human-robot kinematics interaction and bioinspired assistance. The human-robot kinematic coupling
results from the combination of the physical human-machine
interface (pHMI), a kinematic chain of passive degrees of
freedom (pDOF), and several size regulations. All of these
components ensure the device fits properly and the human
and robot joint axes self-align smoothly, thus distributing and
minimizing the transfer of parasitic forces to the user at contact points. The assistive profile matches the biological gravitational angle-torque profile of the upper limbs, providing a
smooth partial gravity compensation that reduces the load on
upper-limb muscles, particularly those in the shoulder,
involved in elevation movements.
MARCH 2020

IEEE ROBOTICS & AUTOMATION MAGAZINE

IEEE Robotics & Automation Magazine - March 2020

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - March 2020