U.S. patent application number 17/558771 was filed with the patent office on 2022-04-14 for wearable joint augmentation system.
This patent application is currently assigned to Dephy, Inc.. The applicant listed for this patent is Dephy, Inc.. Invention is credited to Jean-Francois Duval, Hugh Herr, Luke Mooney.
Application Number | 20220110814 17/558771 |
Document ID | / |
Family ID | 1000006052168 |
Filed Date | 2022-04-14 |
View All Diagrams
United States Patent
Application |
20220110814 |
Kind Code |
A1 |
Mooney; Luke ; et
al. |
April 14, 2022 |
WEARABLE JOINT AUGMENTATION SYSTEM
Abstract
The present disclosure is directed to an autonomous exoskeleton
device that includes one or more actuators, one or more
controllers, one or more sensors with one or more unidirectional
transmissions. The control system includes an exoskeleton member
configured and arranged on a limb of a user; a control device, a
control device connected to the at least one exoskeleton member; an
actuator mechanically connected to the limb of the user; and a
sensor configured and arranged to sense a global angle of the
exoskeleton device relative to the ground. The control device is
configured and arranged to use the global angle to control the
exoskeleton member.
Inventors: |
Mooney; Luke; (Sudbury,
MA) ; Duval; Jean-Francois; (Belmont, MA) ;
Herr; Hugh; (Somerville, MA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Dephy, Inc. |
May |
MA |
US |
|
|
Assignee: |
Dephy, Inc.
Maynard
MA
|
Family ID: |
1000006052168 |
Appl. No.: |
17/558771 |
Filed: |
December 22, 2021 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
16506709 |
Jul 9, 2019 |
11234888 |
|
|
17558771 |
|
|
|
|
62695973 |
Jul 10, 2018 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
A61H 1/0262 20130101;
A61F 2002/704 20130101; A61H 2205/108 20130101; A61H 2201/165
20130101; A61B 5/112 20130101; A61B 2562/0219 20130101; B25J 9/0006
20130101; A61H 2201/5012 20130101; A61F 2002/6827 20130101; A61H
2201/5061 20130101; A61H 2201/5069 20130101; A61H 2205/106
20130101 |
International
Class: |
A61H 1/02 20060101
A61H001/02; B25J 9/00 20060101 B25J009/00 |
Claims
1. A method for measuring motion of a body joint and applying
torque thereto, comprising: providing a control system for a
biomechanical exoskeleton joint; arranging at least one exoskeleton
member on a limb of a user; connecting a control device to the at
least one exoskeleton member; mechanically connecting an actuator
to the limb of the user; actuating the actuator so no forces are
being applied to the exoskeleton joint to provide a zero-torque
control condition, whereby the user does not experience any torque
of the actuator; sensing motion of the limb by at least one sensor;
generating data by the at least one sensor; collecting the data;
analyzing the data; determining an amount of desired torque for the
user; providing information regarding the determined desired torque
to the control device; and applying the desired torque to the
exoskeleton joint via the control device.
2. The method of claim 1, wherein the at least one sensor is a
timer, accelerometer, gyroscope, angle sensor, strain gage,
pressure sensor, force sensor, magnetic field sensor, speed sensor,
optical sensor surface electrode, implanted electrode, implanted
distance measurement sensors, GPS, WIFI transmitter, WIFI receiver,
BLUETOOTH, cellular transmitter, cellular receiver, near field
radio, or a range finder.
3. The method of claim 1, wherein the data collected is one or more
of the following: step time, swing time, stance time, ankle angles
as a function of gait time, shank angles as a function of gait
time, accelerations as a function of gait time, angular velocities,
foot pressure magnitude and location as a function of gait time,
muscle activation magnitude, global position, and gait pattern.
4. The method of claim 1, wherein the limb of the user is a lower
leg.
5. A method for measuring motion of a body joint and applying
torque thereto, comprising: providing a control system for a
biomechanical exoskeleton joint; arranging at least one exoskeleton
member on a limb of a user; connecting a control device to the at
least one exoskeleton member; mechanically connecting a
unidirectional actuator to the limb of the user; actuating the
unidirectional actuator so no forces are being applied to the
exoskeleton joint to provide a zero-torque control condition,
whereby the user does not experience any torque of the actuator;
sensing motion of the limb by at least one sensor; generating data
by the at least one sensor; collecting the data; analyzing the
data; determining an amount of desired torque for the user;
providing information regarding the determined desired torque to
the control device; and applying the desired torque to the
exoskeleton joint via the control device.
6. The method of claim 5, wherein the at least one sensor is a
timer, accelerometer, gyroscope, angle sensor, strain gage,
pressure sensor, force sensor, magnetic field sensor, speed sensor,
optical sensor surface electrode, implanted electrode, implanted
distance measurement sensors, GPS, WIFI transmitter, WIFI receiver,
BLUETOOTH, cellular transmitter, cellular receiver, near field
radio, or a range finder.
7. The method of claim 5, wherein the data collected is one or more
of the following: step time, swing time, stance time, ankle angles
as a function of gait time, shank angles as a function of gait
time, accelerations as a function of gait time, angular velocities,
foot pressure magnitude and location as a function of gait time,
muscle activation magnitude, global position, and gait pattern.
8. The method of claim 5, wherein the limb of the user is a lower
leg.
9. The method of claim 5, wherein the unidirectional actuator is a
winch actuator with a cord that is unwound to a position whereby it
cannot be engaged by the exoskeleton joint to achieve the
zero-torque control condition.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application is a continuation of U.S. Non-Provisional
patent application Ser. No. 16/506,709, filed on Jul. 9, 2019,
which claims priority to earlier filed U.S. Provisional Patent
Application No. 62/695,973, filed on Jul. 10, 2018, the entire
contents of which are incorporated herein by reference.
BACKGROUND OF THE INVENTION
[0002] The invention relates generally to an exoskeleton, a device
worn by a person to augment physical abilities. Exoskeletons can be
considered passive or active. Passive devices do not require an
energy source, such as a battery. Active devices require an energy
source to power electronics and usually one or many actuators. It
is desirable for exoskeletons to be as lightweight as possible,
since the user must carry and move the device along with the body.
It is also desirable for these devices to be capable of providing
large amounts of force, torque and/or power to the human body in
order to assist with motion. These two requirements of low mass and
high force/torque/power are often competing requirements and design
tradeoffs must be made. Furthermore, it is difficult to apply large
forces and torques to the human body. The musculoskeletal system of
the human body is capable of sustaining incredible amounts of
torque and force, but the exterior of the body is not accustomed to
withstanding similar magnitudes of force/torque. Along with being
lightweight, and capable of producing high forces/torques/powers,
exoskeletons should also be comfortable and efficient at
transferring energy to the human. Furthermore, the device should
not interfere with the natural range of motion of the body.
[0003] It is also desirable for active exoskeletons to be energy
efficient and easily controlled. Active exoskeletons require an
energy source to power electronics, sensors and usually actuators.
Typically, batteries are used with electric motors. However,
compressed air can also be used to power pneumatic exoskeletons.
The exoskeleton should be as efficient as possible at converting
the energy source into useful mechanical force/torque/power. Since
the user is often required to also carry the energy source, an
efficient device results in a lighter device, a primary design
objective. Onboard electronics allow designers to control the
exoskeleton, but the device can be mechanically designed to allow
for easier control. For example, active devices with a lower
transmission ratio are often easier to control and back drive.
Output force and torque sensors can also be used to make
controlling easier.
SUMMARY OF THE INVENTION
[0004] The present invention preserves the advantages of prior art
exoskeleton devices. In addition, it provides new advantages not
found in currently available exoskeleton devices and overcomes many
disadvantages of such currently available exoskeleton devices.
[0005] The invention is generally directed to the novel and unique
exoskeleton designs that address the problems associated with known
exoskeleton devices relating to design challenges of device mass,
force/torque/power output, comfort, efficiency and controllability,
and the like.
[0006] The present invention provides an autonomous exoskeleton
that includes one or more actuators, one or more controllers, one
or more sensors with one or more unidirectional transmissions. The
present invention also provides a mechanical joint in parallel with
a biological joint. The exoskeleton device preferably includes an
electric motor and a winch, chain, belt, cam transmission or other
mechanism for providing unidirectional force. Moreover, a
controller, a motor angle sensor, joint angle sensor and/or force
sensor may be provided which, in concert, can be used to calibrate
the controller for the unique movement of the user in an active or
passive mode of the motors. Further still, the motor can be
configured to operate in an active mode, whereby the motor is
imparting energy to the system, and a passive mode in which the
motor is not imparting any, or very little, force to the system.
The motor may be any type of motor but is preferably brushless in
configuration where its diameter is larger than its length. The
present invention can additionally provide for active feedback from
system sensors into the controller to alter motor output during
active use of the exoskeleton. The present invention, in another
aspect, provides for enhanced connection mechanisms between a
user's shoes and the remainder of the exoskeleton.
[0007] It is therefore an object of the present invention to
provide a new and novel exoskeleton device that is compact,
lightweight and inexpensive to manufacture yet is powerful and easy
to control to address the problems associated with prior art
exoskeleton devices.
BRIEF DESCRIPTION OF THE DRAWINGS
[0008] The novel features which are characteristic of the present
invention are set forth in the appended claims. However, the
invention's preferred embodiments, together with further objects
and attendant advantages, will be best understood by reference to
the following detailed description taken in connection with the
accompanying drawings in which:
[0009] FIG. 1 is a perspective view of one exemplary exoskeleton
device;
[0010] FIG. 2 is a reverse perspective view of the exoskeleton
device of FIG. 1;
[0011] FIG. 3 is a reverse perspective view of the exoskeleton
device of the present invention of FIG. 1 with various components
removed for illustration purposes to show attachment of the drive
belt to the drive spool at an ankle joint;
[0012] FIG. 4 is an exploded perspective view of the exoskeleton
device of the present invention of FIG. 1;
[0013] FIGS. 5A and 5B are left and right side views of an
exemplary shoe attachment mechanism;
[0014] FIGS. 6A and 6B are left and right side views of another
exemplary shoe attachment mechanism;
[0015] FIGS. 7A-7D are various views of footplate according to an
exemplary embodiment;
[0016] FIG. 8 is a cross sectional view of a shoe sole mold
according to an exemplary embodiment;
[0017] FIG. 9 is a rear view of a shoe and footplate according to
an exemplary embodiment;
[0018] FIG. 10A-13 show various exemplary attachment configurations
between a footplate and an exoskeleton;
[0019] FIGS. 14A & 14B are side views of an exemplary
alternative shoe embodiment;
[0020] FIGS. 15A-15F show various alternative exemplary attachment
configurations between a footplate and an exoskeleton; and
[0021] FIG. 16 is a perspective view of an exoskeleton system worn
by a user.
DETAILED DESCRIPTION OF THE INVENTION
[0022] Certain exemplary embodiments will now be described to
provide an overall understanding of the principles of the
structure, function, manufacture, and use of the device and methods
disclosed herein. One or more examples of these embodiments are
illustrated in the accompanying drawings. Those skilled in the art
will understand that the devices and methods specifically described
herein and illustrated in the accompanying drawings are
non-limiting exemplary embodiments and that the scope of the
present invention is defined solely by the claims. The features
illustrated or described in connection with one exemplary
embodiment may be combined with the features of other embodiments.
Such modifications and variations are intended to be included
within the scope of the present disclosure. Further, in the present
disclosure, like-numbered components of the embodiments generally
have similar features, and thus within a particular embodiment each
feature of each like-numbered component is not necessarily fully
elaborated upon. Additionally, to the extent that linear or
circular dimensions are used in the description of the disclosed
systems, devices, and methods, such dimensions are not intended to
limit the types of shapes that can be used in conjunction with such
systems, devices, and methods. A person skilled in the art will
recognize that an equivalent to such linear and circular dimensions
can easily be determined for any geometric shape. Further, to the
extent that directional terms like proximal, distal, top, bottom,
up, or down are used, they are not intended to limit the systems,
devices, and methods disclosed herein. A person skilled in the art
will recognize that these terms are merely relative to the system
and device being discussed and are not universal. Moreover, while
certain aspects of the instant disclosure are discussed with
respect to a leg, a shin, a knee, and a foot of a user, the instant
devices and systems can be implemented on any limb of a user.
[0023] Referring first to FIGS. 1-4, details of an exoskeleton 30
is shown. FIGS. 1 and 2 show two different perspective views of the
exoskeleton 30 from the outside. Details of the different
components are shown in FIG. 3 where various components are removed
for illustration purposes. The exoskeleton 30 disclosed herein can
be substantially similar to that which is disclosed in co-owned
U.S. Pat. No. 10,265,195, which is hereby incorporated by reference
in its entirety. The discussion of the structure of the preferred
exoskeleton 30 are provided herein for perspective alone. The
present invention can be used in conjunction with many other,
alternative, exoskeletons which are not discussed herein.
[0024] Referring to FIGS. 1 and 2, the exoskeleton 30 generally
shows a shank tube 24 with a medial ankle joint bearing housing 26
located on the lower end and a medial actuator housing 22 located
at the top thereof. As will be discussed in connection with FIG. 2,
a motor 1, resides in lateral actuator housing 19 that includes a
control electronics cover 20 thereon. Attached to the medial
actuator housing 22 is a calf attachment 11 to secure the upper
portion of the exoskeleton to a calf portion of the user's leg. The
calf attachment 11 preferably includes a shin slide 12 and a shin
guard 13 as well as a shin pad 14 for additional custom adjustment
for better cushioning and comfort for the user. The components of
the calf attachment may be adjusted, as is well known in the art,
to provide a tight but not constricting fit.
[0025] As a result, this attachment to the upper leg of the user
transfers normal forces to the anterior part of the leg, is
lightweight, easy and quick to secure and adjust, can adapt to many
leg sizes and shapes, has minimal bulk to avoid interference with
other pieces of equipment, does not limit range of motion (minimal
medial, anterior and posterior protrusions), can be used over pants
and is comfortable to the user.
[0026] The medial ankle joint bearing housing 26 includes a lateral
ankle joint bearing housing 25 the pivotally receives ankle joint
cross member 5 therein. The free end of the lever arm 4 is fixed to
the joint cross member 5. Therefore, dorsiflexion motion of an
ankle joint causes the ankle joint lever arm 4 to move accordingly,
namely in the direction A shown in FIG. 2. A composite footplate,
generally referenced as 6, has a number of components. It includes
a socket portion 6a, a vertical connector portion 6b and composite
shank 6c. Eversion and inversion of an ankle joint is permitted by
the pivoting action of the rounded free end 6a of composite
footplate 6 within ankle joint cross member 5 in a cylindrical
shaft and socket type pivoting interconnection 6a, namely in
direction B shown in FIGS. 1 and 2.
[0027] A lower free end 4b of lever arm 4 is fixed connected to the
ankle joint cross member 5 while the upper free end 4a of the lever
arm 4 is connected to drive belt 3. The lever arm 4 is preferably
angled upwards at a 50-degree angle so that it does not protrude
beyond the vertical plane at the back of the heel. This angling
also improves the variable transmission profile.
[0028] The drive belt 3 is wound up and unwound about a drive spool
2 driven by motor 1, as shown in FIG. 3, which has medial actuator
housing 22 removed for illustration purposes. With the housing 22
removed in FIG. 3, control electronics 15 and power jack 18 may be
seen residing therein.
[0029] Referring now to FIG. 4, an exploded view of the exoskeleton
30 of the present invention is shown. The ankle motor 1 is mounted
between the medial actuator housing 22 and the lateral actuator
housing 19, namely, within seat 19a of lateral actuator housing 19.
Control electronics 15 are mounted to the exterior surface of the
lateral actuator housing 19 and cover 20 is affixed thereon. A
motor angle sensor magnet mount 16 is provided to carry motor angle
sensor magnet 17. The power jack 18 is mounted to the lateral
actuator housing 19 as well. Drive spool 2 is fixed to motor 1 so
that rotation of motor 1 rotates the drive spool 2. A medial motor
bearing 21 is also provided between the spool 2 and medial actuator
housing 22 for improved smooth operation. The free end 2a extends
clear of inner face 22a of the medial actuator housing with the
free end 3a of drive belt 3 affixed thereto. Thus, as will be
described below, rotation of drive spool 2 causes the drive belt 3
to be wound and out as it is being wrapped and unwound from about
the drive spool 2. The configuration of the drive spool 2 may be
modified, as needed. For example, the diameter, length, profile and
eccentricity of the drive spool 2 may be modified, as needed to
achieve the required winding and unwinding action of the drive belt
3.
[0030] The calf attachment 11 is fastened to the medial actuator
housing 22 by fasteners 11a. The cushioning shin pad 14, shin guard
13 and shin slide 12 are adjustably interconnected with one another
to secure the upper portion of the exoskeleton 30 to the user's
body, such as a leg calf. Other structures and configurations may
alternatively be used to secure the exoskeleton 30, as desired.
[0031] Still referring to FIG. 4, the shank tube 24 includes an
upper shank tube ferrule 23A to enable it to be mounted between the
lateral actuator housing 19 and medial actuator housing 22. A
bottom shank tube ferrule 23B enables the lower portion of the
shank tube 24 to be secured between the lateral ankle joint bearing
housing 25 and medial ankle joint bearing housing 26.
[0032] The ankle joint cross 5 includes a bottom socket 5c and a
pivot member 5b at the top thereof. While the shank tube 24 is
fixed to the lateral ankle joint bearing housing 25 and medial
ankle joint bearing housing 26, the ankle joint cross 5 is
pivotally connected to the lateral ankle joint bearing housing 25
and medial ankle joint bearing housing 26 wherein pivot boss 5b
sits within seat 26a of joint bearing housing 26 via a medial ankle
joint bearing 10. On the opposing side, another pivot boss 5c is
provides that pivotally communicates with lateral ankle joint
bearing housing via lateral ankle joint bearing 9. An ankle angle
joint sensor 7 and ankle angle joint sensor magnet 8 are provided
to sense rotational movement of ankle lever arm 4 relative to the
shank tube 24 and the user's calf position.
[0033] The ankle joint cross also includes a socket 5a to pivotally
receive cylinder 6a of composite footplate 6, which also include a
connector member 6b and composite shank 6c that may receive a sole
of footwear or may be incorporated directly into a sole of footwear
(not shown in FIG. 4), as discussed further below. Ankle joint
cross cap 27 is provided on the free end thereof. As discussed in
FIGS. 1 and 2 above, such a cylinder and socket configuration
permits eversion and inversion of the ankle joint. Eversion and
inversion movement of the ankle joint and, in turn, eversion and
inversion movement of the footwear 32 is permitted due to the
cylinder and socket arrangement movement in the direction of the
arrows shown.
[0034] In some embodiments, a shoe can be specifically designed to
interface with exoskeleton, as discussed in detail below. The
exoskeleton can apply torques around the ankle joint. In general,
the distal structure of the exoskeleton is connected to the shoe
and the proximal structure of the exoskeleton can be connected to
the lower leg shank. In some embodiments the distal structure of
the exoskeleton can be either connected to an existing shoe or
directly integrated into a shoe.
[0035] For example, in one exemplary embodiment, as shown in FIGS.
5A and 5B, the distal structure of the exoskeleton 130 can be
connected to a footplate 106 disposed in a shoe through a series of
removable straps 106a, 106b. A strap 106b may wrap around the toe
of the shoe and another strap 106a may wrap around the heel/arch of
the shoe. The straps can be oriented to resist the torque produced
by the exemplary exoskeleton. A footplate may or may not be
included as part of the attachment mechanism.
[0036] The distal structure may also attach to the foot at a shoe
insert/footplate that goes inside of the shoe, similar to an
orthotic insert. The vertical shoe structure may exit the shoe at
the footplate through an access hole, not shown, in the shoe. The
access hole can include seals that prevent water from the entering
the shoe at exit hole. The vertical shoe structure may also exit
the shoe through the hole of the shoe where a user inserts their
foot. Other attachment strategies include pins and clips, like for
bicycle shoes.
[0037] In one exemplary embodiment, the distal part of the
structure of an exoskeleton 230 can also be directly integrated
into the shoe 235, as shown in FIGS. 6A and 6B. One implementation
can use an exemplary composite foot structure integrated directly
into the sole of a shoe.
[0038] As shown in FIGS. 6A and 6B, a foam upper midsole 238 is
attached to a shoe upper 236. In general, a shoe construction is
shown. For example, as shown, a footplate 206 is attached to the
distal side of the foam sole 238a and a lower foam midsole 238b is
attached to the distal side of the footplate 206. The midsole 238a,
238b material can be other soft conformable materials. The rubber
outsole 240 can be attached to the distal side of the lower foam
midsole 238b. In some embodiments, sensors 250, as shown in FIGS.
7A and 8, can be disposed between the footplate 206 and the outsole
240 to record needed data for the operation of the exoskeleton. For
example, in some embodiments, the sensors 250 can include pressure
sensors and an IMU sensor. The sensors 250 can be wired or wireless
sensors which can communicate with the central controller for the
exoskeleton 230.
[0039] In some embodiments, not shown, an alternative footplate can
be attached directly to a shoe upper. The lower foam midsole can be
attached to the distal side of the footplate and a rubber outsole
can be attached to the distal side of the lower foam midsole. In
another alternative construction, the foam upper midsole can be
attached to a shoe upper, the footplate can be attached to the
distal side of the foam sole, and the lower foam outsole can be
attached to the distal side of the footplate.
[0040] In a further alternative, as shown in FIG. 13, the footplate
206' can be attached to the distal side 238d' of the foam sole 238,
and the rubber outsole 240' can be attached to the distal side of
the footplate 206'.
[0041] When used in an exoskeleton, the footplate can also include
a vertical foot structure 206b, similar to vertical connector
member 6b of FIGS. 1-4, such that the vertical foot structure 206b
protrudes from the bottom of the shoe, proximate an upper surface
of the outsole 238/240, and extends upwards towards the upper
collar of the shoe 235. This vertical foot structure 206b can be
part of the footplate 206, as shown in FIGS. 6A-9 or it may be a
separate structure that is connected to the footplate in either a
rigid, flexible, or jointed manner, as shown in FIGS. 10A-12C. The
vertical foot structure 206b may include padding 207, as shown in
FIG. 9, to protect both the user's body from the structure or to
protect the vertical foot structure from the environment. In one
exemplary embodiment, the footplate 206 can include a foam
protective pad 207 on the lateral side of a laterally protruding
vertical foot structure 206b. The vertical foot structure 206b may
also be connected to the footplate 206 in a manner that is easily
connected and disconnected, as will be discussed further below.
Various fasteners such as pins, magnets, screws, rods and clasps
may be used to quickly connect and disconnect the vertical foot
structure from the footplate. The vertical foot structure 206b may
extend from the footplate on the lateral, posterior and/or medial
sides of the footplate, but the lateral side is the preferred
implementation.
[0042] In general, as shown in FIGS. 7A-7D, the footplate 206 can
be constructed from multiple layers of composite materials 272a-n.
For example, the layers 272a-n can include various carbon fiber,
Kevlar fibers, natural fibers and/or fiberglass sheets are adhered
with epoxy or any other resin or combinations thereof. The
orientation a of the fibers 273, shown in FIG. 7D, can be chosen to
create various levels of stiffness in various directions as is
needed for various performance needs of the user and specific
exoskeleton being used. In the exemplary embodiment, the fibers 273
can be oriented along at +/-45.degree. angle from a longitudinal
axis L of the foot plate. The number of layers 272a-n of the
footplate 206 can additionally be varied throughout the footplate
to create various thicknesses. The geometry of the footplate 206
can be designed to match the shape and curve of the bottom of a
foot of a user. The footplate 206 can be curved upward at the heel
206h and toe 206t, as shown in FIG. 7B. In some embodiments, the
toe of the footplate can be thinner to maintain flexibility.
[0043] Other design considerations can be used to affect the
performance of the footplate 206. For example, layer(s) 272a-n of
unidirectional fibers 273 can be used to increase strength along
certain directions. Further, the layers of woven fibers 273 can be
used to increase torsional strength along certain directions. In a
preferred embodiment, the vertical foot structure 206b and
footplate 206 are a single composite piece. Layers of woven fiber
273 can be used on the vertical foot structure 206b to increase
torsional stiffness about a vertical axis. Layers of unidirectional
fibers 273 can be used in the footplate to increase strength. In
some embodiments, the fibers 273 can also be oriented to run from
the top, posterior edge of the vertical foot structure to the
medial metatarsal joints. In this alternative embodiment, the
fibers 273 can promote efficient load transfer from the vertical
foot structure to the footplate when a plantar-flexion torque is
applied about the ankle joint. In one exemplary embodiment, the
heel 206h of the footplate 206 can be approximately between 1-4 mm
thick and the thickness of the toe area 206t of the footplate can
be approximately 0.3-2 mm thick. The range of thickness can be
approximate, or substantially in the recited range within an
acceptable manufacturing tolerance of +/-0.1 mm. The acceptable
tolerances in thickness can allow for slight variation above or
below the acceptable thicknesses that may occur as a result of
machining or wear.
[0044] In one embodiment the footplate 206, or a part of it, can be
a stack of thin carbon plates, not shown. The stack can be changed
to obtain more or less stiffness or to change how the forces are
transferred. For example, three, or more, layers can provide a
stiffer footplate, where one or two layers would be less stiff. The
stack can be customized at the factory, or by the user to ensure
that the footplate conforms with necessary performance
specifications. The customizations can be user defined or can be
determined with a second device which can scan the geometry of the
user's feet or the device can scan the range of motion of the
user's feet during different situations, such as walking, running,
jumping, lifting, climbing stairs, etc, or combinations thereof. In
some instances, the thin plates can be bonded together. In others
they will not be glued. The stack will be placed in the shoe, under
the insole.
[0045] In some embodiments, as shown in FIG. 8, a footplate 206
having at least one sensor disposed thereon, can be assembled into
a sole 320 of a shoe. The process can include the use of a mold 300
having an upper plate 302 and a lower plate 304. The upper plate
302 and lower plate 304 can come together to form a cavity with a
foot shaped mold 306 to form the upper and mid-soles of the shoe.
The mold 300 can additionally include a structure 310 that is
configured to hold the footplate 206 during the molding process.
The mold can be filled with any desired flowable material.
[0046] The variety of variations on the footplate disclosed herein
can be used with any of the following connection mechanisms
described for connecting the footplate with the lower structure of
the exoskeletons. In some embodiments, plural connections types, or
mechanisms may be used to facilitate the use of a plurality of
exoskeleton types.
[0047] In some embodiments, as shown in FIGS. 10A-12, the vertical
foot structure 206b can be connected to the footplate 206 in a
manner that is easily connected and disconnected. As shown in FIGS.
10A and 10B, the connection mechanism 280 can include at least one
pin 209, on the vertical foot structure 206b, that can be received
in a cutout 286 on the footplate 206, and the vertical foot
structure 206b can be rigidly connected to the footplate 206 by
means of a clasp or collinear hole 284 received in a cutout 288.
The pin 209 can be inserted into the cutout 286 first, then the
clasp can be engaged into the cutout 288 to lock the vertical foot
structure 206b to the footplate at an engagement location 290.
[0048] In an alternative embodiment, as shown in FIG. 11, a
footplate 406 can include one or more horizontal pins 409 that run
in the anterior-posterior direction along the lateral side of the
footplate 406. While similar, but different, reference numerals are
used herein, the remainder of the structure of the exoskeleton and
the shoe can be substantially the same as other embodiments
disclosed herein. Similar to the exoskeleton 30 of FIGS. 1-4, the
medial ankle joint 426 can be actuated, or pivoted by a free end of
the lever arm 404 is fixed to the vertical structure 406b to pivot
the footplate 406 relative to the shank tube 424 about the pivot
point 452. The pins 409 may have a circular or non-circular cross
section. The pins 409 may allow for relative rotation between the
vertical foot piece 406b and the footplate 406 by inserting the pin
409 into the circular pin receiving portions 411, 412 of the
vertical foot piece 406b and the footplate 406, respectively. In
some embodiments, the hinge can be similar to a traditional door
hinge. In FIGS. 12A and 12B, alternative pin shapes 406', 406'' and
hinge shapes 411', 412', 411'', 412'', are shown. The pins 406,
406', 406'' may also be screws. If the pins 406, 406', 406'' are
removeable, then the vertical piece 406b may have a corresponding
hole to receive the pins and connect to the footplate 406. If the
pins 406, 406', 406'' are permanently attached to the footplate,
then the vertical foot structure may use articulated clasps or a
twist-on mechanism to attach to the pins. In one preferred
embodiment, the vertical structure can be inserted and removed from
the footplate when the vertical structure is in one orientation and
is affixed to the footplate when in the worn orientation. For
example, the vertical foot structure 406b can be attached or
removed from the footplate 406 when the vertical foot structure
406b is parallel with the ground. The act of twisting the vertical
foot structure upwards results in it locking into the
footplate.
[0049] In one embodiment, the footplate can have receiving holes
and the vertical foot structure has corresponding pins. The pins of
the vertical foot structure slide into the corresponding footplate
holes in a direction that is orthogonal to the forces exerted by
the exoskeleton. As noted above, the footplate and vertical foot
structure may also have sensors embedded or attached to them. These
sensors may include inertial measurement units, pressure sensors,
strain sensors, force sensors. These sensors may also contain their
required power sources and wireless communication capacities, they
may scavenge energy, or they may be connected via wires to the
control system of the exoskeleton.
[0050] There are various methods to manufacture the footplate, as
discussed above. The shape and thickness of the footplate may be
determined by the direct measurement of a subject's foot, either by
the user (phone app) or by a company employee (phone app, 3D
scanner, mechanical measurement, etc.). The adhesive or materials
in the footplate may also be conformable when heated and shape
stable during operating conditions. A footplate could be heated and
worn by a user to achieve the desired shape. The cooled footplate
would then maintain the desired shape during use. In some
instances, the heat molding process can be done multiple times to
accommodate changes in user's physiology, preferences, or to fit a
different user.
[0051] Another method of manufacturing the footplate can use a
core. A core could be 3D printed or manufactured using another
process. Rapid, digital processes are favored. The composite
laminated would then be adhered to the surface of the core. The
core provides both strength by increasing the area moment of
inertia, along with providing a shape to form the composite layup.
Furthermore, conformable bladders of fluid could be used to press
the curing composite against the core. In some instances, the core
can be dissolved before use, leaving a hollow composite
structure.
[0052] Along with acting as connection point for the vertical foot
structure, the footplate can provide protection and passive energy
storage. The rigidity and strength of the footplate can be used to
protect the bottom of the user's foot from protruding objects, such
as rocks and edges. It can act as a rock plate and as a stabilizer.
The elasticity of the footplate can also be used to store and
release energy during the gait, as shown with respect to FIG.
15F.
[0053] As shown variously in FIGS. 15A and D, the vertical foot
structure 706b may only be attached to the footplate 706 and
actuator (not shown in FIGS. 15A-E), or as in the preferred
embodiment, the vertical foot structure 706b can be attached to a
mechanical joint that mimics the range of motion of the human
ankle, as shown in FIG. 11. Such a mechanical joint can be similar
to the one shown in FIGS. 1-4. In one embodiment, the vertical foot
structure 706b can be connected to the ankle joint structure 724
through a first rotational joint 727 with an axis of rotation A
non-parallel to the plantarflexion/dorsiflexion axis of rotation,
such as the eversion/inversion axis of rotation. The ankle joint
structure 724 can then connected to the leg structure through a
second rotation joint with an axis parallel to the
plantarflexion/dorsiflexion axis of rotation, not shown in FIGS.
15A-F. The first joint 727 may be a low friction joint, or it may
have designed friction. The friction in the first joint may also be
adjustable to resist motion, via tightening of pin 709 disposed in
the hinges 711/712. The first joint 727 may use a plurality of
bearings and have a mechanism to quickly disconnect the vertical
foot structure from the ankle joint structure. The first joint 727
may be a screw 709 that can be tightened or loosed to adjust the
joint friction. The first joint may also be flexural. The first
joint may implement springs 761, 762, as shown in FIG. 15C, or
dampers to affect the torque about the joint in a passive manner.
The first joint 727 may also include a hard stop 763 that limits
the maximum angles of rotation, as shown in FIG. 15B. These
hard-stops may include soft bumpers 764, as shown in FIG. 15E, that
reduce the acceleration when hitting the hard stop. The first joint
may also include sensors to measure forces or angles of
rotation.
[0054] In a preferred embodiment, the second joint is free to
rotate and designed to minimize friction. The second joint may also
implement dampers and or springs during part or all of its range of
motion. For example, springs may be integrated across the second
joint to only impart a force during certain angular ranges. The
second joint also has sensors to measure the angle displacement
between the ankle joint structure and the leg structure. The second
joint may also implement hard-stop features to protect the foot
from being over-extended.
[0055] The ankle joint structure also connects to the actuator to
impart forces between the ankle joint structure and the leg
structure. The actuator is connected to the ankle joint structure
at a point that is non-coincident with either the first joint axis
or the second joint axis of rotation. Forces from the actuator
impart a torque about one or both joints of the ankle joint
structure.
[0056] The leg structure may be attached to the shin or connected
directly to the upper part of the shoe, such as the tongue of the
shoe, as shown in FIGS. 14A and 14B. For example, the shoe 631 can
have an extended shaft 633 that extends upwards towards the knee of
the user. The shoe 631 in this embodiment can have the exoskeleton
630 directly attached thereto through shin attachment mechanism 620
and a similar connection mechanism between the exoskeleton 630 and
the vertical foot structure 606b. While a pin 609 and hinge 611,
612 are shown, any of the aforementioned connection mechanism can
be used. The leg structure may also be attached to the lower leg
through a series of joints that allow for motion in directions
orthogonal to the force created by the actuator.
[0057] Protective pads 540, as shown in FIG. 16, can be integrated
in the leg structure of the exoskeleton 530 to maintain the user's
ability to kneel and crawl. A skid plate may be integrated in the
shoe to make certain bodies of the exoskeleton slip instead of
directly impacting the malleolus.
[0058] As shown in FIG. 15F, the footplate 806 may also include
additional features that allow for the storage and release of
energy. An additional composite leaf spring 810 can be integrated
into the footplate 806 that stores energy during heal strike 800h
and releases the energy at toe off. The spring 810 may look like a
horizontal `Y` with the top of `Y` 810t under the heel 800h and the
bottom of the `Y` 810b under the toe 800t. The arms of the `Y` are
elastic and can be compressed. The elastic footplate 806 may be
manufactured as a single composite piece or as separate pieces that
are adhered in a separate process.
[0059] It is also an important aspect of the present invention to
consider the attachment of an exoskeleton's 530 battery 532. The
battery 532 may be integrated into the main structure 530 (attached
to the structure, integrated in the composite tube, etc.), or into
the shoe 531 (upper part of the shoe, as part of the sole, etc.)
but to decrease distal mass the battery 532 may also be worn on a
different location on the body. In one embodiment, the battery 532
can be worn in a backpack or around the waist, as shown in FIG. 16.
Power cables 535 can connect the battery to the exoskeleton, via a
connection mechanism 560. The central pack 532 may also include
additional computing hardware and the cables 535 may also include
additional data wires. These cables 535 can we worn outside the
pants 570, inside the pants, or integrated into the pants.
Conductive thread or other wearable conductive textiles can be
directly integrated into the clothing to both power and connect the
exoskeleton with the battery. Additional channels can also be sewn
into the clothing to provide a conduit for connecting wires. In
some embodiments, more than one batteries 532 can be used so that
they can be interchanged without disrupting the power to the
exoskeleton 530. One might be mounted at the waist, and each
exoskeleton can have a small battery in it.
[0060] There are various methods for charging the batteries 532, as
shown in FIG. 16, of an exoskeleton, such as the exoskeletons
disclosed herein. The batteries may be removable and charged in a
separate unit. Such a configuration can allow for the rapid
replacement of batteries and ability to reduce down time. A
charging port may be used on the exoskeleton to charge the
batteries while integrated. Wireless charging may also be possible
under certain circumstances. For example, the exoskeleton may
include an inductive coil or other wireless charging mechanism that
can receive power while a user is sitting in a car or at a desk. In
some embodiments, the controller of the exoskeleton can be
programmed such that, upon engaging a wireless charging system, the
actuators are required to be in an inactive state. For example, if
a wireless charger is integrated into a vehicle's seats, or other
structure such as a chair, for charging while the user is driving,
the master controller of the exoskeleton will, advantageously,
ensure that the actuators are in a passive mode so that the
operation of the vehicle is not impeded. For example, if the
actuators were to activate during the operation of a vehicle, the
gas pedal might be pushed too far and the vehicle might undesirably
accelerate causing an accident. In other situations where
non-contact wireless charging occurs, such as non-contact wireless
charging occurring in a warehouse or factory, the actuators of the
exoskeleton may continue to operate in an active mode. In some
embodiments, certain electronic tags, e.g. RFID tags, can instruct
the controller to engage an active or inactive mode of the
actuators depending on the needs of the user in that environment.
For example, an RFID tag in a vehicle may instruct the controller
to disengage the actuators for reasons discussed above. On-board
power electronics can also be used to accept a wide range of
charging voltages, such as 12 V from a car. Alternatively, the
on-board power electronics can be designed to accept any charging
voltage.
[0061] In operation, actuators for an exoskeleton must be efficient
and capable of producing large torques. As discussed above, the
actuators can include an electric motor that uses a transverse-flux
magnetic topology. The motor may also implement advanced materials
such as graphene or nanotube conductors.
[0062] In some embodiments, the exoskeleton can be developed to be
modular in construction. For example, a separate ankle exoskeleton
module and a knee exoskeleton module can be designed independently.
If worn at the same time, the two systems can be both mechanically
and/or electrically attached to transmit forces and/or information
and electric power between them.
[0063] Certain exoskeleton systems can use a transmission that
includes a lever arm to increase the torque at the joint, as shown
in FIGS. 1-4. Such a lever arm may be static, adjustable in length,
and or adjustable in orientation. The adjustability of the lever
arm could be done manually or automatically though additional
actuators. One example is a lever arm that has two configurations:
one for walking, and one for running.
[0064] An exoskeleton may also include mounting points for
temporary parallel actuators. For example, an ankle exoskeleton may
be designed to temporarily accept a parallel damper. The parallel
damper could be used for landing from a fall. The damper may be
reusable or a material that is destroyed after a single use.
[0065] An exoskeleton designed with a unidirectional actuator may
have more than one configurations. That configuration can be the
attachment point of the transmission. One example is a belt that
can be mounted on two different points. In one configuration is
allows powered plantar-flexion, and in the other it allows powered
dorsiflexion.
[0066] The high peak to average current ratio of typical
exoskeleton actuators can be problematic for batteries and for
energy efficiency. Power electronics can be used, in combination
with energy storage (battery, capacitor, inductor) to reduce this
ration while maintaining a constant bus voltage.
[0067] In use, the motor 1 is powered and controlled by the onboard
control electronics 15 and a battery (e.g. 532 of FIG. 16). Angle
of the motor 1 can be measured with the motor angle sensor 16, 17
and the angle of the ankle output joint is preferably measured with
a separate angle joint sensor 7, 8. The sensor can be any type of
sensor, such as an optical encoder, magnetic angle sensor, hall
effect sensor, potentiometer, capacitive sensor, inductive sensor,
or a linear variable differential transformer (LVDT). The various
sensors can be used alone or in any combination.
[0068] In one example, the motor angle sensor 16, 17 and ankle
angle sensor 7, 8 are preferably related when the actuator is
engaged and exerting torque, but independent when the actuator is
not engaged. Thus, during operation, the control electronics 15
controls the take up and pay of the drive belt about the drive
spool. The different components of the exoskeleton 30 are
electronically interconnected to the control electronics 15 so they
may be controlled and monitored as required. For example, the
sensors, motor 1 and power supply, such as a battery (not shown),
are electronically connected to the control electronics 15 whereby
the timing of such take up and pay out of the drive belt 3 via the
motor 1 can be timed or synchronized to the gait of the user, with
the assistance of the sensors, so that the plantar flexion torque
can be applied by the exoskeleton at the appropriate time to use
the work created by that torque to assist in the plantar flexion of
the joint to, in turn, facilitate walking.
[0069] For example, the control electronics 15 can be programmed to
carry out different tasks, such as inertial sensor readings, clock
synchronization between the micro-controllers, serial and PC
communication, non-volatile memory interface, and the like. These
features enable better high-level controllers. For example, a 168
MHz Cortex-M4F STM32F427 controller may be used to carry out
computing cycles available for executing high-level algorithms
concerning control of the exoskeleton 30. Field Programmable Gate
Arrays (FPGA), Complex Programmable Logic Devices (CPLD),
Application Specific Integrated Circuits (ASIC), and Graphical
Processing Units (GPU) may also be used for hardware math
acceleration and control. This makes it also possible to use
machine learning techniques in real time.
[0070] In operation the exoskeletons use one or many computing
elements (microprocessor, micro-controller, programmable logic) and
a selection of sensors to control the various actuators or single
actuator. In the instant disclosure, there can be at least two
fundamental modes that the exoskeletons operates under: a zero
torque mode and an active mode.
[0071] During zero torque mode the exoskeleton is programmed such
that it does not apply any torque about the augmented joint or is
controlled to apply as close to zero torque as possible. The
present disclosure considers that there may be various levels of
zero torque. If the device had zero mass and no artificial joints,
then the user would truly feel zero torque since there would be no
device. If the device has mass but no artificial joints or
actuators, then the user would only feel the inertia of the added
mass. This could be perceived as an additional torque about the
joint. If the device has an artificial joint, then the user will
feel a drag torque because of the physical joint. This torque
results from friction in the joint and/or bearings. These torques
are likely imperceivable by the user, due to their small
magnitudes, just like the forces created by the upper part of a
tall boot or by elastic socks or pants (ex: compression socks,
leggings, etc.)
[0072] Substantial torques about the joint are felt when the device
includes an actuator. For example, assume a device that has a motor
directly connected to the joint. The user will experience the
inertia of the motor rotor as an external torque that is
proportional to the angular acceleration. Furthermore, the user
will experience any friction in the motor as a resistive torque. If
current is able to flow through the leads of the motor, then the
user may also experience a drag torque produced by the back EMF of
the motor. The motor could be controlled to attempt to maintain
zero torque on the joint. This type of controller can mitigate the
effects of motor drag torques, but it is exceedingly difficult to
develop a controller that can anticipate the motion of the user and
compensate for the inertial effects of the motors. The controller
must know the current acceleration of the joint and appropriately
control the motor to match the joint position, velocity and
acceleration. Sensor delay, computational delay, and filtering
delay all contribute to system delays that make it impossible to
know the exact dynamics state of the joint and apply the
appropriate control.
[0073] An actuator that is capable of disengaging from the joint
can exert a true zero-torque against the joint. One example would
be a motor that has a clutch between the motor and the joint. When
the clutch disengages the motor from the joint, the user would not
experience the torques due to the motor. However, this requires
additional mechanical complexity and mass. Another option is to
implement a unidirectional actuator that is only capable of
applying forces in a single direction. For example, an actuator
that uses a cord to exert forces across the joint can exert a
torque in one direction but is unable to exert a torque in the
opposite direction. Simply, a cord can only pull and not push. A
simple controller can be programed to maintain slack in the cord
and ensure that no forces are being applied to the joints.
Alternatively, the cord can be kept in a position that is beyond
the angular range of the joint and thus can never be engaged. A cam
and follower are another example of a unidirectional actuator.
[0074] A zero-torque mode can important for a number of reasons.
One such reason is for observing the natural behavior of a user
while wearing an exoskeleton without being impeded by forces
generated by the actuators or the weight of the exoskeleton itself.
During zero-torque mode, the exoskeleton can use its sensors and
onboard microprocessors to measure and analyze the unimpeded motion
of a user. For example, a lower limb exoskeleton can measure gait
parameters of the user while walking during zero-torque control.
Important bio-mechanical parameters can include joint angles,
velocities and accelerations, limb accelerations and angular
velocities, and the timing of these parameters with respect to
periods of the gait cycle.
[0075] During an active mode, the exoskeleton can have periods of
applying torque and periods of applying zero-torque. The
exoskeleton controller may also continue to measure the user's gait
parameters during the active mode to continuously adjust output of
the actuators during use. The algorithm that determines the
application of torque may use parameters measured during the
zero-torque mode, the active mode, or a combination of the two.
[0076] An ankle exoskeleton, like the one disclosed herein, can use
various strategies to apply torque about the ankle. The present
discussion is made with respect to the ankle; however, it is
understood that the present exoskeleton, controller, and program
can be used with any bio-mechanical joint in the body. The
exoskeleton can be configured to apply torque about the ankle as a
function of the measured ankle angle. The ankle angle may be
measured with an angle sensor and/or inertial measurement units.
The ankle angle estimate may be adjusted as a function of the
torque applied.
[0077] Alternatively, or additionally, the exoskeleton actuators
can apply torque about the ankle as a function of the measured
shank angle. The shank angle may be measured with various angle
sensors and/or various inertial measurement units. The shank angle
may be adjusted as a function of the torque applied. It is
understood that the term "shank angle," as used herein, is the
global angle of the shank's longitudinal axis with respect to the
floor/earth or a relative change in shank angle over a period of
time. In effect, the torque of the individual actuators can be
changed as a function of the measured shank angle. For example, the
shank angle can be set to zero upon the measurement of the heel
striking the ground, then measure a change in shank angle with an
integrated gyroscope located on the shank itself.
[0078] In general, the exoskeleton can additionally, or
alternatively, apply torque about the ankle as a function of time.
The present algorithm can also use a combination of exoskeleton
torque profile strategies.
[0079] One exemplary method for measuring the shank angle of user
wearing an exoskeleton is to attach a microprocessor to the
exoskeleton and a gyroscope sensor to the shank of the user. The
gyroscope sensor can be the only sensor on the exoskeleton, or the
gyroscope can be used in combination with other sensors on the
exoskeleton. The gyroscope measures the angular velocity of the
shank. The control algorithm can receive the angular velocity and
estimates the shank angle. The algorithm is as follows:
[0080] 1) The angular velocity is measured to be above a
threshold;
[0081] 2) The angle is set to zero;
[0082] 3) The swing time is set to zero;
[0083] 4) The angle is computed as the time integral of the angular
velocity;
[0084] 5) The swing time is incremented;
[0085] 6) The angular velocity is measured to be below the
threshold;
[0086] 7) The peak swing angle is set to be the current estimate of
the angle;
[0087] 8) The swing time is set at the current time;
[0088] 9) The shank angle is set to zero;
[0089] 10) The stance time is set to zero;
[0090] 11) The shank angle is computed as the time integral of the
angular velocity; and
[0091] 12) The stance time is incremented.
The shank angle can then be transmitted to the exoskeleton
controller as an input variable for a torque output function for
the actuator. The algorithm can also be narratively described as
follows, the angular velocity of the limb can be continuously
received by the controller, typically at a fixed frequency. During
each control loop, the controller can use logic to determine which
state it should be in. For example, if the angular velocity is
below a predetermined threshold, and the controller is not
currently in the swing state, then the controller enters the swing
state. The swing state can be controlled, for example if the
controller was not in the swing state during the previous cycle,
then 1) the swing timer is set to 0 and the swing angle is set to a
value (0 in one embodiment, but it does not have to be and can be
change over time); 2) increment the swing timer; 3) increase the
swing angle by integrating angular velocity, 4) if the angular
velocity is above a certain threshold and if the swing timer and/or
swing angle are above certain thresholds, then the swing angle is
saved and used by controller and the swing timer is saved and used
by controller, and the controller does not re-enter the swing state
in the next cycle; 5) if the previous conditions are not met, then
re-enter the swing state in the next cycle.
[0092] In another, or additional, exemplary method, a user can
first wear the exoskeleton and power it on. The user may use a
separate device, such as a phone, tablet, or computer, to manually
adjust the parameters of the exoskeleton. These parameters may
include: gender; age; height; weight; carried weight; desired
walking speed; desired running speed; limb lengths; type of
terrain; proficiency; desired gait (jump, squat, crawl, etc.);
energy harvesting mode; known gate pathologies or injuries; and
other preference: e.g. speed increase vs calorie saving.
[0093] The user may also select pre-recorded profiles that describe
control strategies. The device can additionally or alternatively
perform a calibration routine. Calibration may include sensor
zeroing, such as angle sensors and inertial measurement sensors.
The calibration may also include the synchronization of an actuator
motor and joint angle sensor. The calibration may include having
the user do specific motions to measure his range of motion and
maximum joint speeds. If the user has not selected a pre-recorded
profile, then the user may perform a series of motions while the
exoskeleton is in a zero-torque control condition, as discussed
above. If the actuator is a unidirectional actuator, such as a
winch actuator, then the zero-torque control may be achieved by
unwinding the cord to a position that cannot be engaged by the
ankle joint. The zero-torque control condition may also be achieved
by the controlling the cord to maintain a certain level of slack by
controlling the motor position as a function of the ankle
position.
[0094] While the exoskeleton is in a zero-torque mode, the
controller can measure various gait parameters using various
sensors. The sensors may include: timers; accelerometers;
gyroscopes; angle sensors; strain gages; pressure sensors; force
sensors; magnetic field sensors; speed sensors; optical sensors;
surface electrodes; implanted electrodes; implanted distance
measurement sensors; GPS; WIFI transmitter/receiver; BLUETOOTH;
cellular transmitter/receiver; near field radios; and range
finders.
[0095] The parameters measured by the various sensors may include
various combinations of: step time; swing time; stance time; ankle
angles as a function of gait time; shank angles as a function of
gait time; accelerations as a function of gait time; angular
velocities; foot pressure magnitude and location as a function of
gait time; muscle activation magnitude; and global position. The
instant system is additionally capable of deriving certain
measurements as a function of the measured parameters. For example,
the derived measurements can include: measurements as a function of
another measurement; ankle angle during range of shank angle;
muscle activation during range of ankle angle; accelerations during
period of time; peak measurements as a function of another
measurement; peak ankle angle during a period of time; peak angular
velocity during swing; peak pressure during a range of ankle
measurements; and peak muscle activation during a period of
time.
[0096] After a period of time, or number of gait events, the
exoskeleton may enter an active mode. The exoskeleton can monitor
the gait and can estimates user gait patterns. The exoskeleton
controller may use statistical models to predict the level of
confidence that a certain gait is detected. Regressions may be used
to predict certain gait parameters as a function of one or many
sensor measurements. For example, during the zero-torque mode, the
controller may compare the peak plantarflexion ankle angle and the
step time of a single step. A regression model has been developed
to correlate step time with the peak plantarflexion ankle angle.
This correlation would can additionally measure the strength of the
correlation. If the strength is high enough, the controller could
use the step time of a step to predict the peak plantar flexion
angle of the next step. This strategy could be used with any
combination of measured and or derived parameters.
[0097] If the exoskeleton controller does not recognize a specific
gait pattern or has low confidence in the regression model, the
controller may choose to enter a zero-torque mode at any time. The
zero-torque mode may be entered in a gradual manner to reduce the
effects of abrupt transitions. The exoskeleton controller may also
choose to enter the zero-torque control mode if an error is
detected such as a missing sensor, broken actuator or nearly
depleted battery.
[0098] If more than one exoskeleton is worn at a time (i.e. a right
and left exoskeleton) the multiple exoskeletons may communicate
with each other wirelessly or over wire. The multiple exoskeletons
can additionally communicate with a central controller, which can
coordinate controls of both of the exoskeletons. The multiple
exoskeletons can share sensory information and parameters in real
time with each other and use these measured and derived parameters
to inform the join control of the system as a whole. For example,
the ankle angle of the left leg could be used to inform the
controller of the right leg. Calculations may also be shared across
multiple systems to reduce computational burden. The two individual
devices can do their share of the same calculation, or they can use
different algorithms. The algorithm with the highest confidence
rating can then be used to control the combination of the two
devices. The two devices can additionally, or alternatively, run
the exact same algorithm as a form of redundancy.
[0099] In some embodiments, loop optimizations can be performed to
optimize the performance of the instant exoskeleton, or other
exoskeletons. The control algorithm can apply slight perturbations
to the system (change power level, add delay or anticipate action,
change gains, etc.) and use onboard and off-board sensors to
measure the impact on the user. One example is a master algorithm
that tries to minimize the user's heart rate frequency at a given
energy. User feedback, obtained via an application or by pressing
on a button, can be used as part of the optimization strategy.
[0100] The exoskeleton can be part of a wired or wireless network
of device. These devices can include the one or more of following:
a mobile device (e.g., a phone, a tablet, etc.); Other exoskeleton
controllers; a computer or a laptop; a cellular network;
wireless/wired sensors; heart rate monitors; temperature sensors;
oxygen consumption sensors; muscle activation sensors; lab
equipment; and gym devices (treadmill, stair climber, elliptical
machine, etc.).
[0101] The exoskeleton control systems can receive updates over a
network. For example, a mobile device may be able to send an update
to the exoskeleton to inform the controller. The exoskeleton data
may be uploaded to a central database, where multiple user data is
collected, combined and analyzed. The exoskeleton data may also be
stored on a decentralized network. These data may be used to inform
generic controllers. A mobile device may also be able to provide
information to the user about the state of the exoskeleton, such
as: battery level; missing or broken sensors; broken actuators;
missing communications; errors; power/energy/torque provided to the
user; number of steps; energy saved; distance traveled; map of
travel; fitness metrics; and total time use and other usage
statistics.
[0102] The described exoskeleton can be worn on one leg or on both
legs in a bilateral configuration. When the exoskeleton is worn in
a bilateral configuration, the two exoskeletons can communicate
with wires or wireless communication protocols to share state
information for purposes of control and telemetry.
[0103] It would be appreciated by those skilled in the art that
various changes and modifications can be made to the illustrated
embodiments without departing from the spirit of the present
invention. All such modifications and changes are intended to be
covered by the appended claims.
* * * * *