user does not need to pay attention to control it, instead the
wearer moves naturally as the system is able to follow the
wearer movements like a ’shadow’.
EMG signals contain rich information that can be used
to control and drive an exoskeleton robot and similar manmachine
devices in both rehabilitation and assistance [8]. The
wearable robot Hybrid Assistive Limb (HAL) developed by
Sankai et al. [3], is comprised of four actuators for hip and
knee assistance and uses feedback from both EMG and force
sensors. Similarly, Li et al. [5][4] and Fleischer et al. [2][9]
use EMG signals to control an exoskeleton.
However, in contrast to laboratory environments, where
the temperature and humidity are controlled in an indoor
setting, industrial environments present a big challenge for
monitoring technologies that are glued or fasten to the
exoskeleton’s wearer body.
EMG is considered as being overly complex and subject
to external conditions for use in industrial exoskeletons to
modulate its assistance. Thus, the wearer may sweat then
affect the EMG reading or the connectors of the EMG
sensors might detach which make this option not reliable.
A similar approach is to fasten the monitoring sensors to
the wearer, however, this can become invasive and due to
movements the sensors might shift positions or get loose,
making them not a reliable solution for industrial settings.
The robotic assistance of an industrial exoskeleton may
alternatively be regulated according to external conditions,
such as the weight of the object to be handled. In this
case, force sensors are integrated in the wearer’s footwear or
within the plantar area of the exoskeleton where the wearer
is standing on. Hence, from the force sensors it is possible
to detect the variation of planar pressure which enables the
exoskeleton’s controller to trigger the assistive force.
Ekso Bionics [10] is a commercial application using force
sensors for physical rehabilitation. It includes force sensors
in the footwear of the patient to analyze the pattern of the
gait, in order to provide assistance for walking practice. The
Ekso Bionics system places high-dense force sensors in the
footwear by covering the whole plantar area, in order to
ensure that the sensors receive all the pressure force. Due to
this extensive placement of sensors, the system is required
to adjust to individual foot shape and size making difficult to
be portable to common footwear other than their own special
footwear integrated in their exoskeleton system. On the other
hand, due to the massive use of sensors, the Ekso Bionics
system is among the highest cost applications in its field, by
pricing over 100,000 USD [11].
An exoskeleton is an external structural mechanism with
joints and links corresponding to those of the human body
[1]. Their main use is to assist the wearer by boosting
their strength and endurance for different purposes, such as
military, medical or civilian use.
It is a common practice in laboratory that the exoskeleton’s
wearer controls the system by indirect measures, such as
electromyography (EMG) [2][3][4][5] or force/torque measurement
generated from interaction [6][7]. This means that
the wearer does not directly control the exoskeleton with a
joystick as common cases for robots or mechanical systems.
The reason behind is that the hands and in some cases the full
body of the wearer are expected to be free. In this context,
the ideal exoskeleton should be able to infer the intention
of the user by indirect and reliable measures in which the