SOFT ROBOTICS

Volume 00, Number 00, 2018

ª Mary Ann Liebert, Inc.
DOI: 10.1089/soro.2017.0090

ORIGINAL ARTICLE

Precharged Pneumatic Soft Actuators

and Their Applications to Untethered Soft Robots
Yunquan Li,1 Yonghua Chen,1 Tao Ren,2 Yingtian Li,1 and Shiu hong Choi3
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.

Abstract

The past decade has witnessed tremendous progress in soft robotics. Unlike most pneumatic-based methods, we
present a new approach to soft robot design based on precharged pneumatics (PCP). We propose a PCP soft
bending actuator, which is actuated by precharged air pressure and retracted by inextensible tendons. By pulling
or releasing the tendons, the air pressure in the soft actuator is modulated, and hence, its bending angle. The
tendons serve in a way similar to pressure-regulating valves that are used in typical pneumatic systems. The
linear motion of tendons is transduced into complex motion via the prepressurized bent soft actuator. Fur-
thermore, since a PCP actuator does not need any gas supply, complicated pneumatic control systems used in
traditional soft robotics are eliminated. This facilitates the development of compact untethered autonomous soft
robots for various applications. Both theoretical modeling and experimental validation have been conducted on
a sample PCP soft actuator design. A fully untethered autonomous quadrupedal soft robot and a soft gripper
have been developed to demonstrate the superiority of the proposed approach over traditional pneumatic-driven
soft robots.

Keywords: precharged pneumatics, soft actuator, controllable actuation, tendon control, soft robotic gripper,
untethered soft robot

Introduction pump.13–16 Instead of continuous bending deformation, Ne-

miroski et al. presented a compliant joint actuated by air and
retracted by an elastic tendon.17 However, the compliant joint
S oft robots are commonly constructed from soft
and compliant materials. They are generally actuated by
means of pneumatic power, which offers the benefits of high
needs tethers for external air supply. It is extremely difficult
to design and develop a soft robot that is large and strong
power-to-weight ratio, ready availability, and cost-effective enough to carry a self-contained pneumatic system, including
materials and fabrication.1,2 batteries, motors, air compressors, circuit boards, valves, and
This article focuses primarily on soft robots actuated by tubes.18
soft pneumatic actuators (SPA), which are characterized by There have been attempts18–22 to make soft robots function
their inflatable, low-weight elastic membrane and inherent autonomously. One way is to combine pneumatics and
environmental compliance.3,4 When actuated by compressed combustion.19–21 Since the combination can reduce the usage
gas, the soft actuator chambers expand in the directions with of compressed gas, only a miniature compressor is carried
low stiffness, and hence, the desired motions such as bend- onboard. However, this approach can achieve only very
ing,5,6 twisting,5,7 extension/contraction,8 and combined simple motions such as jumping. Controllability is also a big
motions.9 Modeling, improvement, and advanced fabrication problem. To initiate a jump, the robot inflates a subset of its
of pneumatically powered soft actuators are reported in legs by compressed gas for jumping directional control. On
Refs.10–13 A wide range of possible applications of such soft combustion, the robot belly balloons out abruptly, pushing
robots are explored in Refs.14–16 Many previous soft robots against the ground and propelling the robot into the air. The
are actuated pneumatically using compressed gas/air through complicated ignition sequence, consisting of fuel delivery,
flexible tethers, normally tubes, from a stationary air mixing, and sparking, causes a time delay between jumps.
1
Department of Mechanical Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong.
2
Robotics Research Center, Xihua University, Chengdu, Sichuan, China.
3
Department of Industrial and Manufacturing Systems Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong.

1
 2 LI ET AL.
To eliminate tethering to any hard robotic control systems
and power sources, a soft robot with microfluidic control and
powered by gas from chemical reaction is reported in Ref.22
The gas generated from fuel combustion inflates the fluidic
networks downstream of the reaction sites, resulting in actua-
tion. However, microfluidic logic can only provide very simple
control strategies. Given the large strain of the soft robot,
clogging of the microfluidic channels poses a great challenge.
Furthermore, the force that the soft robot can exert on the
environment is very small, limiting its practical applications.
To increase the load capability of untethered soft robots,
Trolley et al. developed a large (0.65 m long) soft robot,
actuated by pneumatic power.18 The large size is made pos-
sible by fabricating the robot using composite materials
comprising elastomer, polyaramid fabrics, and hollow glass
microspheres. Even at this large size, the robot still has dif-
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.
ficulties in encapsulating all the necessary power supplies
(batteries and compressed gas cylinders) and other compo-
nents (valves, electric motor-driven air pump, circuit board)
needed for autonomous operations, as shown in Supple-
mentary Figure S1 (Supplementary Data are available online
at www.liebertpub.com/soro). Furthermore, the robot’s lo-
comotion speed is very slow, as slow as 2.0 m/h in an un-
dulating gait limited by the small flow rate of compressed air.
To increase the flow rate, larger compressors, valves, and
tubing are needed, which may cripple the robot.
In this research, we report a novel approach to pump-less
and untethered autonomous soft robotics. Although the pro-
posed approach adopts the commonly practiced methods for
FIG. 1. Principle of the proposed
PCP actuator. (A) Comparison be-
tween a traditional soft pneumatic
actuator (1) and the proposed PCP
actuator (2). (B) PCP actuator bend-
ing and its controllable parameters: F
(Pulling Force), V (Tendon Pulling/
Releasing Speed), and D (Tendon
Pulling/Releasing Displacement). (C)
The control system of a PCP actu-
ator: 0 is defined as the initial state
and 1 as the fully actuated state.
PCP, precharged pneumatics. Color
images available online at www
.liebertpub.com/soro
the soft actuator design (Fig. 1A(1)), it does not need any
connection tubes for air supply and release. Instead, the air
chamber is completely enclosed, like a basketball or a tire,
leaving only a small orifice (check-valve) for air charge
(Fig. 1A(2)). As such, once the actuator is charged, it can be
used for months before the next charge is required. We call
this precharged pneumatics (PCP) soft actuator, inspired by
air gun design. In operating an air gun, a cocking drive is
pulled back to compress air in a pump. By pulling the trigger,
air pressure is released and shot out the pellet by pneumatic
power. When the PCP soft actuator is charged, the tendons
(cables) pull back the already bent PCP actuator to a desired
position. By pulling or releasing the tendons, the actuator’s
air pressure is modulated, and hence, its deformation, as
shown in Figure 1B, where the actuator exhibits different
degrees of bending due to different tendon lengths. At each of
the actuator position, its air pressure is different. The PCP
soft actuator has three controllable variables that can be
controlled one at a time: the tendon pulling force F, the
tendon pulling/releasing speed V, and the tendon pulling/
releasing displacement D. Such controllability allows the
actuator to be used for various applications. The control
system is also much easier and simpler, as shown in
Figure 1C, where the complicated pneumatic system (com-
pressor, valves, and tubes) is completely eliminated. This
greatly facilitates the development of untethered autonomous
soft robots, as shown in the later parts of the article.
Tendon control is much easier and more accurate than
pneumatic control. This is because gas is compressible, and
 PCP SOFT ACTUATORS AND UNTETHERED SOFT ROBOTS
thus, a pneumatic system is regarded a nonlinear system.23
Since the gas pressure can hardly be controlled quickly, it is
difficult to control the deformation of a soft actuator accu-
rately.24,25 Moreover, when the tendons are long and their
winding requires more sliding contact, the tendon driving
mechanism may have nonlinearity and insensitivity problems
when the load is large.26,27 In contrast, the proposed PCP soft
actuator is driven by precharged air, not by the tendons.
Actuator design and analytical modeling
A sample PCP soft actuator has been designed for analysis
of its characteristics, which is validated by a set of experiments
using a prototype fabricated from the design. Figure 2 shows
the 3D schematic drawing and details of the sample design.
The actuator is composed of two parts, the soft actuator and the
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.
actuator base. The soft actuator is a classic fiber-reinforced soft
actuator design.1 Its cross section is semicircle with a radius
16 mm (Fig. 2A). Three silicone rubber pads are molded onto
the back of the soft actuator for the tendon guide. We use the
Kevlar cord as the pulling tendons because of its extremely
high tensile strength-to-weight ratio.
Each of the silicone rubber pads has two through holes
(Fig. 2B). Silicone rubber tubes (with higher hardness than
3
that of the pads) are installed inside the holes to reduce the
friction between the tendons and the silicone rubber pads, as
well as to protect the pads from being cut by the cord. Through
the tendon guide inside the actuator base, the two tendons
converge to a single point for better control. After assembly,
the total length of the actuator is about 140 mm (Fig. 2C).
We use a pedal air pump connected with a gas needle to
inflate the actuator (Fig. 2D). After inflating the actuator, the
air pump is removed and the actuator is at the bending state.
We design a valve pattern that is quite similar to the bas-
ketball check-way valve structure, and mold it with silicone
rubber (Fig. 2E). Figure 2F shows the check-valve cross-
sectional view and the working principle. The valve channel
has different diameters, which gradually decrease from the
bottom (3 mm) to the top (0.5 mm). A mechanical clamper is
designed for clipping the smaller channel completely air-tight
(Fig. 2F). When a gas needle is inserted into the valve, the
small channel will stretch a bit for pumping air into the
chamber. When the actuator bends to a desirable angle, the gas
needle is pulled out and the clamper squeezes the valve air-
tight, trapping the air in the chamber. Due to the small hardness
(SHORE A 10D) of silicone rubber used for making the pro-
totype, the air pressure that the valve can sustain is about 0.10
Mpa, and the actuator bends about 100 under this pressure.
FIG. 2. Design of a proto-
type PCP actuator. (A) Front
view of the actuator, (B)
overall actuator design, (C)
prototype of the PCP actua-
tor, (D) precharged actuator,
(E) check-valve placement,
and (F) cross-sectional view
of the check-valve. Color
images available online at
www.liebertpub.com/soro
 4 LI ET AL.

Given a PCP bending actuator design and a precharged air
pressure, we can model the bending angle, the tendon pulling
force, and the corresponding bending force.28 Supplementary
Figure S2 shows the soft actuator model and its parameter
definitions. A detailed analytical modeling is given in the
Supplementary Data.
Experimental validation
We build a test platform shown in Figure 3A to measure
the actuator’s characteristics. First, we use the platform to
measure the tendon pulling force and the actuator bending
angle against the tendon displacement. Two linear guides
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.
controlled by step motors are fixed vertically on the table. A
force gauge is installed on the sliding block with a hook head.
The actuator tendons are anchored on the hook head (Fig. 3A).
The force gauge can be controlled to move forward and
backward at different speeds. We use a high-resolution
camera to record the actuator trajectory. The tendon force
can be read from the force gauge. To analyze the actuator
trajectory, we use the software Kinovea to measure the
bending angle and the displacement of the tendon from
the video record.
The experiment starts from the fully actuated state of the
actuator (Fig. 3A). The force gauge is pulled at 10 mm/step
for nine steps at a speed of 6.0 mm/s. For each step, there is a

FIG. 3. Experimental validation. (A) Experimental setup, bending angle at fully actuated state: h  98 . (B) Actuator
bending trajectory. (C) Actuator end pad tilt angle measurement. (D) Pulling force changes tendon displacement. (E)
Relationship between the tendon displacement D and the bending angle h. (F) Relationship between the tendon dis-
placement D and tendon pulling force F. (G) Bending force measuring experimental schematic and setup. (H) Relationship
between pulling force F and bending force FB. Color images available online at www.liebertpub.com/soro
 PCP SOFT ACTUATORS AND UNTETHERED SOFT ROBOTS
pause of 2 s for the actuator to stabilize. The pulling forces are
measured and the moving trajectories are video recorded.
When the actuator reaches the initial state (straight), the force
gauge releases the tendon to the fully actuated state. The
entire process is repeated six times.
Bending trajectory and actuator end pad tilt. The bending
trajectory of the actuator is shown in Figure 3B. The red dash
line shows the trajectory of the endpoint of the actuator
during the tendon pulling and releasing processes. The tra-
jectory shows good consistency. It should be noted that while
the actuator is controlled by the tendons, the actuation is
powered by the precharged air. This is quite different from
the direct tendon-driven actuators.29 The body of the pro-
posed PCP soft actuator exhibits a continuous and smooth
bending curve.
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.
When analyzing the actuator trajectory, we notice that
there is a tilt angle at the end pad when the actuator is pulled
toward its initial state. The tilt is not so notable when the
change in the bending angle is within Dh  60 . However,
the tilt becomes large when the bending angle h changes from
30 to 0 (Fig. 3B) because the end pad bears most of the
tendon pulling force. Thus, the end silicone rubber pad de-
forms a lot during the pulling process. As shown in Figure 3B,
the tilt angle c is defined as the angle between the pad edge
and the line along the actuator bottom layer (tangent line to
the deformation segment when actuator bends). During a
complete actuating process, the tilt angle can vary as large as
30 ~35 .
Measurement of the errors induced by pad tilting. The
pad tilt will no doubt induce an extra pulling force and an
extra pulling distance compared with the analytical model.
Therefore, we design an experiment to measure these errors.
We precharge the actuator to the same pressure as in the
previous experiment (0.1 Mpa) and then use a force gauge
and a flat barrier to confine the actuator at its initial state
(bending angle 0) (Fig. 3C). We pull the actuator tip for a
certain distance at steps of 1 mm and then release the actuator
tip slowly to its start position, such that the tilt angle changes
from c1 to c2 . We measure the pulling force using another
force gauge. The result shows that the actuator pad tilt in-
duces a significant error compared with the analytical model.
When there is a change of 30 in the tilt angle, the pulling
distance can be as much as 35 mm, while the largest pulling
force can be as large as 13 N. We plot the relationship be-
tween the pulling force and the pulling displacement during
the pulling and releasing processes (Fig. 3D). It should be
noted that the gradient of the curve during releasing process is
smaller than that of the pulling process. The release of the
first 5 mm witnesses a rapid descent. This phenomenon is
induced by the elastic hysteresis when silicone rubber is
stretched and released.30
Bending angle and pulling force against tendon displace-
ment. The relationship between the actuator bending angle
and the tendon displacement is plotted based on both ex-
perimental data and the theoretic model (in the Supplemen-
tary Data) as in Figure 3E. It can be seen that there is an error
region between the experimental data and the analytical data.
This is caused by the tilt of the end pad. Based on the ex-
perimental result, the relationship between the bending angle
5
and the tendon displacement is linear. As a result, the bending
angle of the PCP actuator can be linearly controlled by the
tendon displacement.
The relationship between the tendon pulling force F and
the displacement D is shown in Figure 3F. The pulling force
is measured when the PCP actuator is pulled from the fully
actuated state to the initial state, and the releasing force is
measured when the PCP actuator is actuated from the initial
state to the fully actuated state. The pulling force increases
linearly against the tendon displacement during the pulling
process. When the tendon changes from pulling to releasing
(D changes from 90 to 80 mm), the tendon force curve shows
a rapid decline. The rest of the curve (D changes from 80 to
0 mm) in the releasing process remains a steady gradient.
More discussions are provided in the Supplementary Data.
Bending force versus pulling force. The bending force FB
that the actuator can provide is measured based on the setup
in Figure 3G. The relationship between the pulling force
F and the bending force FB based on the experimental data
and the theoretical modeling is shown in Figure 3H. The
largest bending force measured is *5.4 N. More discussion
on the relationship between the two parameters is provided in
the Supplementary Data.
Sample applications
Due to the simple design and controllability of the pro-
posed PCP soft actuators, they can be used for more advanced
soft robotic applications. In this research, we have designed
and constructed a soft robotic gripper with three PCP actu-
ators, and an untethered quadruped soft robot with four PCP
actuators. Both can demonstrate superior performance over
traditional pneumatic soft robots.
Soft gripper. Soft pneumatic bending actuators have been
applied to design many soft robot grippers or soft robot
hands.12,14,31 Wei et al.12 developed a variable stiffness soft
gripper that keeps the compliance property of soft robotics
and modulates the actuator’s stiffness by up to 10 times.
Deimel and Brock14 developed the RBO Hand 2, which is a
completely soft robotic hand with five PneuFlex actuators.
Their design demonstrates both good compliance and dex-
terous performance. Galloway et al.31 developed a soft
gripper that can delicately manipulate and sample fragile
species on the deep reef. Zhou et al.32 mentioned that soft
actuator’s compliance reduces the dependency on force-
feedback control and prevents damage to the object in many
cases. Soft robotic grippers/hands have advantages of grasp
compliance and manipulating delicate objects.
However, to the best of the authors’ knowledge, most
previous soft robotic grippers/hands can be controlled only
under two states: initial state and actuated state. Due to the
nonlinearities of rubber material property and pneumatic
pressure control, precision control of the bending speed and
the bending angle has rarely been discussed. This is indeed a
drawback for most soft pneumatic grippers. As discussed and
analyzed in the previous section, for our PCP actuator, the
bending speed and the bending angle can be easily controlled
by controlling the tendon, which is much easier than con-
trolling air pressure.
 6 LI ET AL.

Another major advantage of our proposed gripper is that
the PCP actuators are precharged and can be easily controlled
by pulling and releasing the tendons. Our gripper eliminates
the need of air supply, and hence, the bulky and heavy air
pump, for its applications. Compared with air source,
electrical power source is more compact, efficient, easily
available, and easy to set up. Therefore, our gripper is more
versatile for various soft robotic applications.
The gripper prototype is shown in Figure 4A. It consists of
a gripper base and three PCP actuators. The three actuators
are installed with two on one side and one on the other side. In
Figure 4B, the PCP’s tendons are wound to the same shaft at
different positions, while the shaft rotates to roll/release the
tendons. The shaft is actuated by a DC motor through a pair of
gears. Based on the pulling force required for controlling the
PCPs, we can select a motor with the right power and torque.
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.
width W1 ¼ 40 mm, gripper open width W2 ¼ 80 mm, and
finger length l ¼ 120 mm.
When the shaft rotates clockwise, the tendons will be
rolled onto the shaft until the PCP actuators become straight,
which is the initial state (Fig. 4C). It should be noted that if
we keep on rotating the tendon shaft, the PCPs can bend
further outward (Fig. 4D) to increase the gripper open width
(W2¢ > 80 mm). At the fully actuated state, the tendons are
completely released, returning the actuators to the natural
bending state without any extra power input (Fig. 4E).
We test the gripper’s performance by grasping different
objects with different sizes, shapes, and mechanical proper-
ties. The gripper grasps a water bottle (500 g), a piece of tofu
(260 g), and a roll of sticky tape (D ¼ 100 mm > W2 ). When
grasping the water bottle (Fig. 4F), like other soft grippers,
the gripper just needs to approach the bottle from a proper

The DC motor is controlled by a controller that can control
the rotation speed and direction. Thus, the actuators’ bending
speed and angle can be controlled accordingly. As shown in
Figure 4B, the basic configuration parameters of the gripper
are as follows: finger assembled angle a ¼ 80 , gripper base
position and fully release the tendons. The largest payload
that our soft gripper can hold is about 600 g. Grasping tofu
and sticky tape requires special control processes. Here we
focus on describing the grasping procedure of tofu and
the tape.

FIG. 4. Prototype soft robotic gripper. (A) The prototype gripper. (B) Schematic diagram of tendon driving mechanism.
(C) Initial state of the gripper. (D) Gripper fingers bending outward. (E) Actuated state of the gripper. (F) Grasp a bottle of
water (weight 500 g). (G) Grasp fragile object Tofu or bean curd (weight 260 g). (H) Grasp a roll of tape (diameter 100 mm).
Color images available online at www.liebertpub.com/soro
 PCP SOFT ACTUATORS AND UNTETHERED SOFT ROBOTS
Tofu, or bean curd, is very fragile. Grasping tofu can
therefore test the gripper’s controllability. When approaching
the tofu, the PCP actuator is controlled to move slowly to
result in a very gentle touch. As force sensors are not used in
the grasp tests, the grasp experiment is guided by human
observation. After several trials, we have identified appro-
priate values for two control variables in tofu grasping: ten-
don releasing speed at below 10 mm/s and tofu indent depth
(5 mm) when the fingers have contact with it. When these two
values are observed, the sample tofu can be easily grasped
without damage, as in Figure 4G.
We further test the actuator by grasping a tape roll with a
diameter D ¼ 100 mm, which is a little bit larger than the
gripper’s initial open width W2 ¼ 80 mm. The grasp may be
unstable as only the fingertips of the gripper are in contact
with the tape. To increase the contact area, the tendons are
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.
pulled to open the gripper further, as shown in Figure 4D.
The gripper then approaches the tape roll and wraps around
it for a stable grasp, as shown in Figure 4H. This demon-
strates that the gripper’s initial state can be changed for
different applications.
The grasping processes of objects in Figure 4 are recorded
and shown in the Supplementary Videos S1–S6.
A fully untethered autonomous quadrupedal soft robot. We
design and fabricate an untethered autonomous quadrupedal
robot with four PCP actuators as limbs to further demonstrate
the actuators’ applications. The robot prototype is shown in
Figure 5A. Its design schematic drawing is shown in Figure 5B.
7
For simplicity, the prototype has only two motors. Actuator
1 and Actuator 2 are controlled by one servomotor, while
Actuator 3 and Actuator 4 by another servo motor. The total
length of the robot is 200 mm and its weight is about 1.1 kg.
The robot body contains a mobile power (10000mAh Mi
Power Bank), an Arduino Mega board, and two servomo-
tors (MG995 servomotor). The servomotors rotate clock-
wise and counter-clockwise in the range of 180. The initial
installed states of the four actuators are different (Fig. 5B),
that is, Actuator 1 and Actuator 4 are at fully bent state,
while Actuator 2 and Actuator 3 are at the initial straight
state.
The walking gait of the robot is shown in Figure 5C(1–6).
The diagonal actuators are always at the same actuation
state. When Actuator 1 and Actuator 4 are bent, Actuator 2
and Actuator 3 are pulled to the initial state, and vice versa
(Fig. 5C(1) and (4)). The friction coefficient between sili-
cone rubber and wooded desk is quite large. Therefore,
when two actuators are bending backward, the reaction
forces push the robot to lean forward (Fig. 5C(2) and (5)). At
the same time, the other two actuators are pulled to the
straight state, sweeping through the ground and moving
forward along with the entire robot body. Sellotape is stuck
to the bottom of the actuator to decrease the friction
(Fig. 5B). The robot can move forward at an average speed
of 5 mm/s through periodically alternating movements. This
speed is much faster than that reported in Ref.18 In fact, the
walking speed can be further increased by using faster servo
motors.
FIG. 5. Untethered autono-
mous quadrupedal soft robot.
(A) Quadrupedal robot proto-
type. (B) Schematic design
diagram and initial assembled
state. (C (1–6)) Walking gait of
the robot. Color images avail-
able online at www.liebertpub
.com/soro
 8 LI ET AL.
Discussion and Conclusion
The degrees of freedom (DOFs) of a traditional robot are
defined by its number of joints. Soft robots have infinite
DOFs and thus show great advantages in flexibility and
conformability in comparison with traditional robots. How-
ever, such advantages pose several challenges and limita-
tions. The use of materials with large deformation properties
results in very complex modeling of soft actuators and un-
stable dynamic performance (vibration, low stiffness, and
unpredictable locomotion) compared to antagonistically ac-
tuated drives. Another limitation of soft actuators, especially
for SPA, is the need to connect to a heavy air pump through a
tether, which greatly hinders its mobility.
In this research, we have presented a PCP soft actuator
controlled by tendons. The basic actuator design follows the
fiber-reinforced soft pneumatic actuator. Major design
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.
changes are in the following two aspects: (1) replacement of
the air inlet with a check-valve and (2) adding tendons to pull
back the actuator to its original position before precharged
with air. Once charged, the air pump is not required. Ex-
periments and experiences have shown that the check-valve
sealing is very effective (basket balls and tires can last for
months before next charge). The bending (or actuating)
process of the PCP actuator is controlled by pulling or re-
leasing the tendons that are controlled by servo motors.
Controlling a servo motor is much easier and more accu-
rate than controlling air pressure and flow rate. We can
control the pulling speed, the pulling displacement, and the
pulling force of the PCP actuator relatively easily. These
variables are directly associated with the soft actuator’s
bending speed, bending angle, and bending force. This ac-
tuator and its control system can be regarded as a continu-
ously deformation nonlinear system constrained by a tendon
that can be linearly controlled. We made a three-fingered
gripper prototype and a quadrupedal autonomous robot with
the PCPs. Through experiments, we found that the actuator
exhibits the advantages of both soft robots and tendon-driven
robots. The actuator shows great compatibility and control-
lability. It can not only bend slowly and gently to grasp
delicate and fragile objects such as tofu but can also bend fast
and provide enough power to drive a walking robot.
Although our current PCP actuator design shows good
performance, there is room for improvement. First, the ac-
tuator size and appearance can be redesigned and optimized.
The current design of the tendon guide support is made rel-
atively sturdy to ensure certain strength. Therefore, the entire
actuator looks bulky and clumsy, impeding its mobility and
flexibility. The tendon guide can be redesigned and made
with stronger materials to reduce the size and enhance the
practicality of the actuator. Second, the hardness of the sili-
cone rubber used for the check-valve is quite small (10D,
Shore A), limiting the precharged air pressure to below
0.1 MPa and hence a maximum bending angle of about 98. A
more reliable check-valve design and the use of better ma-
terials can contribute to a larger precharged air pressure, and
hence, a larger bending angle and longer operation life. Fi-
nally, although the actuator has eliminated the need for air
pumps and valves, a servomotor and battery are still required.
These parts are heavy as well. We are currently investigating
other types of tendons, such as shape memory alloys or shape
memory polymers.
Acknowledgment
This work is funded by a basic research grant from the
University of Hong Kong. A Chinese patent application with
registration number 201710668524.1 is pending approval.
Author Disclosure Statement
No competing financial interests exist.
References
1. Polygerinos P, Wang Z, Overvelde JTB, Galloway KC,
Wood RJ, Bertoldi K, et al. Modeling of soft fiber-
reinforced bending actuators. IEEE Trans Robotics 2015;
31:778–789.
2. Trivedi D, Rahn CD, Kier WM, Walker ID. Soft robotics:
biological inspiration, state of the art, and future research.
Appl Bionics Biomech 2008;5:99–117.
3. Laschi C, Mazzolai B, Cianchetti M. Soft robotics: tech-
nologies and systems pushing the boundaries of robot
abilities. Sci Robotics 2016;1:1–11.
4. Sun Y, Song YS, Paik J. Characterization of silicone
rubber based soft pneumatic actuators. In IROS: Interna-
tional Conference on Intelligent Robots and Systems,
Tokyo, Japan, 2013, pp. 4446–4453.
5. Ilievski F, Mazzeo AD, Shepherd RF, Chen X, Whitesides
GM. Soft robotics for chemistry. Angew Chem Int Ed Engl
2011;50:1890–1895.
6. Suzumori K, Endo S, Kanda T, Kato N, Suzuki H. A
bending pneumatic rubber actuator realizing soft-bodied
manta swimming robot. In ICRA: International Con-
ference on Robotics and Automation, Roma, Italy, 2007,
pp. 4975–4980.
7. Martinez RV, Fish CR, Chen X, Whitesides GM. Elasto-
meric origami: programmable paper-elastomer composites
as pneumatic actuators. Adv Funct Mater 2012;22:1376–
1384.
8. Tondu B, Lopez P. Modeling and control of McKibben
artificial muscle robot actuators. IEEE Control Syst Mag
2000;20:15–38.
9. Connolly Y, Polygerinos P, Walsh CJ, Bertoldi K. Me-
chanical programming of soft actuators by varying fiber
angle. Soft Robotics 2015;2:26–37.
10. Moseley P, Florez JM, Sonar HA, Agarwal G, Curtin W,
Paik CJ. Modeling, design, and development of soft
pneumatic actuators with finite element method. Adv Eng
Mater 2016;18:978–988.
11. Yap HK, Ng HY, Yeow CH. High-force soft printable
pneumatics for soft robotic applications. Soft Robotics
2016;3:144–158.
12. Wei Y, Chen YH, Ren T, Chen W, Yan CX, Yang Y, et al.
A novel, variable stiffness robotic gripper based on inte-
grated soft actuating and particle jamming. Soft Robotics
2016;3:134–143.
13. Mosadegh B, Polygerinos P, Keplinger C, Wennstedt S,
Shepherd RF, Gupta U, et al. Pneumatic networks for soft
robotics that actuate rapidly. Adv Funct Mater 2014;24:
2163–2170.
14. Deimel R, Brock O. A novel type of compliant and un-
deractuated robotic hand for dexterous grasping. Int J Rob
Res 2016;35:161–185.
15. Polygerinos P, Galloway KC, Savage E, Herman M, Donnell
KO, Walsh CJ. Soft robotic glove for hand rehabilitation and
task specific training. In ICRA: International Conference on
 PCP SOFT ACTUATORS AND UNTETHERED SOFT ROBOTS
Robotics and Automation, Seattle, USA, 2015, pp. 2913–
2919.
16. Polygerinos P, Lyne S, Wang Z, Nicolini LF, Mosadegh B,
Whitesides GM, et al. Towards a soft pneumatic glove for
hand rehabilitation. In: IROS: International Conference on
Intelligent Robots and Systems, Tokyo, Japan, 2013,
pp. 1512–1517.
17. Nemiroski A, Shevchenko YY, Stokes AA, Unal B, Ainla
A, Albert S, et al. Arthrobots. Soft Robotics 2017;4:183–
190.
18. Trolley MT, Shepherd RF, Mosadegh B, Galloway KC,
Wehner M, Karpelson M, et al. A resilient, untethered soft
robot. Soft Robotics 2014;1:213–223.
19. Bartlett NW, Tolley MT, Overvelde JTB, Weaver JC,
Mosadegh B, Bertoldi K, et al. A 3D-printed, functionally
graded soft robot powered by combustion. Science 2015;
349:161–165.
Downloaded by SUNY Stony Brook package(NERL) from www.liebertpub.com at 06/20/18. For personal use only.
20. Tolley MT, Shepherd RF, Karpelson M, Bartlett NW,
Galloway KC, Wehner M, et al. An untethered jumping
soft robot. In: IROS: International Conference on In-
telligent Robots and Systems, Chicago, 2014, pp. 561–566.
21. Loepfe M, Schumacher CM, Lustenberger UB, Stark WJ.
An untethered, jumping roly-poly soft robot driven by
combustion. Soft Robotics 2015;2:33–41.
22. Wehner M, Truby RL, Fitzerald DJ, Mosadegh B, White-
sides GM, Lewis JA, et al. An integrated design and fabri-
cation strategy for entirely soft autonomous robots. Nature
2016;536:451–455.
23. Bobrow JE, McDonell BW. Modeling, identification, and
control of a pneumatically actuated, force controllable ro-
bot. IEEE Trans Robotics 1998;14:732–742.
24. Caldwell DG, Medrano-Cerda GA, Goodwin M. Control of
Pneumatic Muscle. IEEE Control Syst 1995;15:40–48.
25. Andrikopoulos G, Nikolakopoulos G, Arvanitakis I, Man-
esis S. Piecewise affine modeling and constrained optimal
9
control for a pneumatic artificial muscle. IEEE Trans Ind
Electron 2014;61:904–916.
26. In H, Lee D, Cho KJ. Investigation of friction character-
istics of a tendon driven wearable robotic hand. In Inter-
national Conference on Control, Automation and Systems,
Gyeonggi-do, Korea, 2010, pp. 568–573.
27. Palli G, Borghesan G, Melchiorri C. Tendon-based
transmission systems for robotic devices: models and
control algorithms. In ICRA: IEEE International Con-
ference on Robotics and Automation, Kobe, Japan, 2009,
pp. 4063–4068.
28. Ogden RW, Mineola NY. Non-Linear Elastic Deforma-
tions. New York: Dover Publications, 1997.
29. Manti M, Hassan T, Passetti G, D’Elia N, Laschi C, Cian-
chetti M. A bioinspired soft robotic gripper for adaptable and
effective grasping. Soft Robotics 2015;2:107–116.
30. Love AEH. Treatise on the Mathematical Theory of Elas-
ticity. Cambridge, UK: Cambridge University Press, 1927.
31. Galloway KC, Becker KP, Phillips B, Kirby J, Licht S,
Tchernov D, et al. Soft robotic grippers for biological
sampling on deep reefs. Soft Robotics 2016;3:23–33.
32. Zhou X, Majidi C, O’Reilly OM. Soft hands: an analysis of
some gripping mechanisms in soft robot design. Int J Solids
Struct 2015;64:155–165.
Address correspondence to:
Yonghua Chen
Department of Mechanical Engineering
The University of Hong Kong
Pokfulam Road
Hong Kong 852
Hong Kong
E-mail: yhchen@hku.hk