In
general, grip strength has been seen to gradually decline between 60 to
75 years of age- this decline more drastically noted among men [1].
Additionally, approximately 795,000 Americans suffer from a stroke, the
leading cause of serious long-term disability, per year, reducing
mobility, including upper limb dysfunction, in over half of stroke
victims age 65 and older [2].
Upper limb dysfunction, including decreased grip strength and/or
diminished ability to hold objects is also prevalent in populations with
carpal tunnel syndrome [3].
Robotic
exoskeleton devices can be primarily designed to augment user strength
in order to assist with activities of daily living (ADLs), or as
rehabilitative devices that are used under the guidance of a physical
therapist to help patients regain greater functionality of damaged
joints and/or muscles [4].
Assistive exoskeletons for the hand can be grouped according to how the
augmenting forces enhance the concentric movement of the digits.
Devices have been designed to apply the augmenting forces to the dorsal
aspect of the fingers via mechanical linkages [5,6,7,8] or fabric-based pneumatic bladders [9]. A ventral approach has also been used, where pseudo-tendons applied tension that is transmitted to the digits through soft [10] or hard exoskeleton structures [11]. Heo, et al. [12] and Bos, et al. [13] have both published comprehensive listings and reviews of exoskeleton devices for the hand.
Regardless
of the technique used to apply the augmenting force, for an assistive
device to function, finger movement or another indication of the user’s
intent to move must be sensed and transformed into a signal that
controls the application of the assistive forces. Ideally, there needs
to be a consistent coordination between the device and the user that
results in a coupling of the human hand and the augmenting system,
allowing the robotic device to consistently provide assistance as needed
through the detection and amplification of the user’s effort. Some
grip-assistive devices, however, have pre-programed algorithms with
which users do not initiate by intent to move. These types of devices,
such as the HERO Grip Glove [14]
move the user’s hand through gripping and/or pinching patterns that
allow for a set force production, which is then augmented by a user’s
own strength. Devices by Yap et al. and Polygerinos et al.
operate in a similar fashion, where the user shows intent to move, and
the device then moves through a pre-determined motion without any
subsequent input from a user [15, 16]. Such devices can both be used for hand motion training with the guidance of a physical therapist, as well as assist in ADLs.
Hand
exoskeleton designs vary in overall weight, complexity, and cost. In
attempts to provide the full range of motion of the human hand to the
user, most of these devices have become both bulky and complex, and due
to this are restricted to a single functional activity- either
hand-opening or pinching. These exoskeleton devices often use a single
motor or driving feature to assist multiple fingers [14], such as with Yoo et al.’s design which used one motor to drive three fingers [17] and Gasser et al.’s design which uses two motors to control four fingers [18]. Alternatively, some devices actively assist fewer than all five fingers [19], for example, Pu et al.’s, Nycz et al.’s, and Gasser et al.’s designs exclude the thumb [6, 18, 20].
Devices that allow for more degrees of freedom and independently assist
all five digits become exceedingly cost prohibitive as more joints,
motors, and custom electrical components become necessary [5, 21].
These additional motors and therefore batteries will also make the
device heavier and potentially tethered to a power source dependent on
the current draw [9, 22].
A
previous design from our laboratory used machined aluminum segments to
construct exoskeleton digits with a desktop computer-based control
system tethered to the device [23, 24].
In order to reduce both the manufacturing cost and time of these
previous prototypes as well as the weight, the most recent exoskeletons
were designed to be constructed with 3-D printed thermoplastics.
Furthermore, a minicomputer-based control system replaced the desktop
computer and associated data acquisition hardware, which provided a
further reduction in cost, weight, and complexity, and also allowed for
greater freedom of movement [11, 25].
Even though exoskeletons with 1 or 2 fingers are simpler to implement,
most ADLs require at least 3 fingers to be assisted by an exoskeleton [26].
Exoskeletons with 3 or 4 fingers could assist with most ADLs, the
realism for the user would decrease as the number of fingers decrease [26].
Additionally, having fewer fingers limits the grasping positions the
user can make with their hand, as well as limit the objects the user can
lift. For example, a 4 or 3-fingered exoskeleton could assist a user in
picking up objects of uniform shapes (e.g., cup, reusable water bottle)
but not objects that are oddly shapes or have varying thicknesses
(e.g., cell phone, wine glass). Increasing the number of independently
controlled exoskeleton digits would allow for the control to lift
objects such as this and increase the mobility to the point where it
feels natural to the user. For training and rehabilitation devices
specifically, being able to independently control each finger is
imperative for re-developing muscle and flexibility in every finger. All
these reasons listed are why most modern powered exoskeletons for the
human hand use a five-fingered design despite the added complexity and
weight [15, 16, 22, 27,28,29], and why we have decided to move from a three-fingered design [11, 25, 30] to a five-fingered design [31].
The
main objective of this study was to design and produce a wearable
powered exoskeleton for the human hand to improve structural stability
of the fingers while also augmenting pinching and grasping efforts, and
to validate that the device augments both the user’s pinching and
grasping efforts and ability to perform ADLs by evaluating healthy human
subjects. The exoskeleton device should be user friendly, allow for
individual finger movement, and be cost optimized. This device aims to
not compromise cost and weight for individual, independent movement of
all five fingers. To be user friendly, the device must be able to
incorporate a range of sizes that users may experience on a daily basis,
as well as have a minimal user interface, and be easily donned and
doffed. Additionally, the device must be portable and easily carried,
and the batteries should last multiple hours. For cost optimization,
electronic components must be commercially available, and the device
should be modular such that broken parts are able to be replaced as
necessary. Additionally, the modularity of the design must be such that
different sized pieces are able be added and removed for users of
differing size in the future. The exoskeleton structure was designed
using CAD (computer aided design) software to enclose all five fingers
of the right hand and was 3-D printed in ABS plastic. Each finger’s
movement efforts were individually monitored and proportionally
augmented via the microcontroller-based control scheme, linear
actuators, and rigid exoskeleton structure.
More at link.
