ORIGINAL RESEARCH

published: 13 May 2020

doi: 10.3389/fbioe.2020.00461

Soft Pneumatic Gripper With a

Tendon-Driven Soft Origami Pump
Yeunhee Kim and Youngsu Cha*
Center for Intelligent & Interactive Robotics, Korea Institute of Science and Technology, Seoul, South Korea

In this study, we propose a soft pneumatic gripper that uses a tendon-driven soft origami
pump. The gripper consists of three pneumatic soft actuators that are controlled by
a tendon-driven origami pump. An external air compressor that supplies air to the
pneumatic actuator is replaced by an origami pump. The soft actuator is composed
of silicone (Ecoflex 00-30) with a chamber-based structure, which is fabricated using a
mold, and the origami pump is fabricated by folding a Kresling patterned polypropylene
film. In addition, we conduct a series of experiments to evaluate the performance of
the pneumatic actuator with a tendon-driven origami pump. Specifically, movement
characteristics, frequency response, blocking force, and the relation between bending
angle and pressure are analyzed from the results of the experiments. Furthermore,
Edited by: we understand the entire operation mechanism from the deformation of the origami
Seung Tae Choi, pump to bending through pressure. Finally, we demonstrate the grasping of objects with
Chung-Ang University, South Korea
diverse shapes and materials, and indicate the feasibility of the pneumatic gripper as an
Reviewed by:
Matteo Cianchetti, independent module without an external compressor.
Sant’Anna School of Advanced
Keywords: biomimetics, origami, soft gripper, soft robotics, soft pump
Studies, Italy
Zicai Zhu,
Xi’an Jiaotong University, China
Hugo Rodrigue, 1. INTRODUCTION
Sungkyunkwan University,
South Korea Soft robotics has recently received considerable attention in the field of robotics. Soft robots, such
*Correspondence:
as locomotive robots (Chan et al., 2012; Tolley et al., 2014; Wang et al., 2014), biomimetic robots
Youngsu Cha (Laschi et al., 2009; Marchese et al., 2014; Snell-Rood, 2016; Lu et al., 2017; Della Santina et al.,
givemong@kist.re.kr 2019), origami robots (Onal et al., 2012; Paez et al., 2016; Salerno et al., 2016; Firouzeh and Paik,
2017; Li et al., 2017; Lee and Rodrigue, 2019), micro robots (Li et al., 2016; Palagi et al., 2016), and
Specialty section: soft manipulator, utilize the flexibility of soft materials (Hughes et al., 2016) in various applications.
This article was submitted to The precious concept source for soft robotics corresponds to biomimetics. It mimics the
Bionics and Biomimetics, movement and morphological characteristics of animals and plants (Laschi et al., 2009; Marchese
a section of the journal
et al., 2014; Wang et al., 2014; Snell-Rood, 2016; Lu et al., 2017; Della Santina et al., 2019). It is
Frontiers in Bioengineering and
also inspired by motions and structures of the human body (Lu et al., 2017; Della Santina et al.,
Biotechnology
2019). A Human-like robotic manipulator is an extremely important and necessary research area.
Received: 30 January 2020
Various robotic manipulators are developed due to the explosive demand for robotic hands in
Accepted: 21 April 2020
manufacturing. They can perform various tasks in several unpredictable situations (Hughes et al.,
Published: 13 May 2020
2016). In contrast to conventional manipulators with a rigid body (Butterfaß et al., 2001; Wojtara
Citation:
et al., 2005), soft manipulators are actuated by various methods, such as hydraulic (Mitchell et al.,
Kim Y and Cha Y (2020) Soft
Pneumatic Gripper With a
2019; Park et al., 2020) and pneumatic (Deimel and Brock, 2016; Jittungboonya and Maneewarn,
Tendon-Driven Soft Origami Pump. 2019), and smart materials, such as ionic polymer metal composite (IPMC) (Carrico and Leang,
Front. Bioeng. Biotechnol. 8:461. 2017; Bhattacharya et al., 2019; Roy et al., 2019), shape memory alloy (SMA) (Rodrigue et al., 2017;
doi: 10.3389/fbioe.2020.00461 Lee et al., 2019), and electroactive polymer (Kofod et al., 2007; Taghavi et al., 2018).

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 1 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

In this study, we propose a soft pneumatic gripper with a involves pouring the silicone solution (Ecoflex 00-30, Smooth-
novel tendon-driven soft origami pump. Specifically, we utilize On, Inc.) in 1/3 of the mold and curing at room temperature
an origami pump as an air supplier for the pneumatic gripper for 4 h. Subsequently, a polyethylene terephthalate (PET) film
instead of typical air compressors. Origami structures are adapted with a thickness of 100 µm is attached to the cured silicone. The
in various fields such as soft robots (Onal et al., 2012), actuators PET film is inserted as a backbone to increase the stiffness of
(Firouzeh and Paik, 2017), and artificial muscles (Li et al., 2017; the actuator and to maintain its shape. Additionally, the silicone
Lee and Rodrigue, 2019) with various materials, such as paper solution is poured such that it fills the entire mold and is cured
(Paez et al., 2016), film (Onal et al., 2012; Li et al., 2017), under the same condition.
and silicone (Sun et al., 2013). Origami can easily be patterned The posterior finger part is fabricated using a 3D printed
and folded in 2D planes to make 3D structures (Demaine and mold for the chamber-based structure (Figure 1C) (Ilievski et al.,
O’Rourke, 2007; Lang, 2011). We design an origami pump 2011; Deimel and Brock, 2016; Galloway et al., 2016). The mold
with a Kresling pattern, which can be folded in the form of is designed using a 3D CAD program (SOLIDWORKS 2015,
a 3D cylinder shape from 2D patterned triangles (Kobayashi Dassault Systèmes Corp.) and manufactured by a 3D printer
et al., 1998; Mahadevan and Rica, 2005; Kresling, 2008; Jianguo (ProJet HD3500 Plus, 3D Systems, Inc.). The build and support
et al., 2015). The benefits of the origami pump include fast materials used in the 3D printer include VisiJet M3 Crystal and
reaction speed, light weight, and reliable features. Also, the VisiJet S300, respectively. Afterward, silicone solution is poured
origami pump can facilitate modular pneumatic actuator through into the 3D printed mold and cured at room temperature for 4
independent configuration. h. After curing, both silicone parts are combined with a silicone
To control the origami pump, we select the tendon-driven bond (Sil-Poxy, Smooth-On, Inc.) to prevent air leakage and
method. The tendon-driven mechanism is a commonly used maintain flexibility. The total length (L) of the soft pneumatic
method for conventional robot hands (Hong et al., 2018) and soft actuator is 70 mm.
grippers (Lee et al., 2019). In our case, the tendon is connected The origami pump consists of Kapton tape (thickness 60
between the origami pump and a motor. We control the cylinder µm) and polypropylene (PP) film (thickness 200 µm). To make
height of the origami pump via the tendon. The tendon controls the folded cylindrical shape, we select the Kresling pattern
the origami pump with a sealed air and it can make the (Kobayashi et al., 1998; Mahadevan and Rica, 2005; Kresling,
pump to have a small operating range than typical pumps. 2008; Jianguo et al., 2015), which can be folded in the form of
In addition, we experimentally and theoretically analyze the a cylinder shape from patterned triangles as an origami. The PP
actuation performance of a finger component in the pneumatic film is patterned by the laser cutter and utilized as a substrate of
gripper. Specifically, we examine the movement characteristics, the origami structure. Each piece of the 2D patterned PP film is
frequency response, blocking force, and the relation between attached between two Kapton tapes and rolled into a cylindrical
pressure and bending angle. Additionally, we demonstrate the shape, as shown in Figure 1D. The original height (H) of the
grasping performance of the soft gripper with diverse objects. origami pump is 40 mm.
This paper is organized as follows: In section 2, we describe The air channel component is inserted between the soft
the design of the soft finger module with a soft origami pneumatic actuator and origami pump. The air channel
pump, including the fabrication method, operation principle, component consists of a 3D printed part and an acrylic channel.
and experimental setup. In section 3, we show and analyze Additionally, it is connected with the tendon attachment for
experimental results of the soft pneumatic finger. Specifically, the tendon-driven operation of the origami pump. The tendon
the relations between input from the origami pump, pressure, attachment is a component that allows the pump to compress
and performance of pneumatic actuator are investigated and upwards through motor operation.
compared with that of the theoretical model. Additionally, the Finally, we add a 3D printed bottom part under the pump that
frequency response of the soft pneumatic actuator is analyzed is attached to seal and connect the tendons for pump operation.
in the section. In addition, we describe the design of the soft The bottom part is attached with the silicone bond to avoid air
gripper consisting of three finger modules, and demonstrate the leakage. After the fabrication process for the soft finger module,
grasping performance with various objects. The conclusions are four tendons (0.6 mm 304 grade S/S wire Rope, COSMO WIRE
summarized in section 4. Co.,Ltd) are installed to the air channel, bottom of pump, and
a motor (MX-28AT, ROBOTIS Co.,Ltd). To guide the tendon
path, we insert the motor-driven part and tendon-guiding part,
2. MATERIALS AND METHODS as shown in Figure 2A. The motor-driven part is manufactured
by a 3D printer and connected to the motor. In addition, the
2.1. Fabrication Method tendon-guiding part is positioned between the pump and motor
The soft finger module consists of a soft pneumatic actuator, air to change the direction of the tendons (Figure 2A). It is also
channel component, and origami pump (Figure 1A). Specifically, utilized for the balancing operation of the origami pump by
the soft pneumatic actuator is composed of two components: the the motor.
anterior finger part and posterior finger part. Figure 1B presents
the fabrication process of the anterior finger part. The mold is 2.2. Operation Principle
a combination of two acrylic bodies manufactured by a laser The operating mechanisms are classified into finger bending
cutter (Epilog laser MINI 18, Epilog laser Corp.). The first step and finger release. The finger bending procedure is divided into

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 2 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

FIGURE 1 | Fabrication process of a soft pneumatic actuator and an origami pump: (A) Soft finger module. (B) Fabrication process of the anterior finger part. (C)
Fabrication process of the posterior finger part. (D) Fabrication process of the origami pump.

FIGURE 2 | Operation mechanism of a soft finger module with the tendon-driven process. (A) Finger module with the motor. (B) Tendon-driven process with
clockwise rotation of the motor. (C) Pump compression corresponding to tendon-driven process. (D) Actuator expansion procedure via air injection from pump
compression. (E) Tendon-release process with counterclockwise rotation of the motor. (F) Pump release via the elastic characteristic of the origami pump. (G)
Actuator release procedure via air extraction of the origami pump.

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 3 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

FIGURE 4 | Experimental setup to measure of (A) optical data from the finger
module operation and (B) output pressure from the origami pump.

record the actuation video using a camera (DSC-RX10M3, Sony

FIGURE 3 | Photograph of soft finger module pump with parameters. Corp.) at a rate of 30 fps. Additionally, we use a post-processing
tracking program (ProAnalyst Motion Analysis Software, Xcitex,
Inc.) to extract the deformation data from the video. The
experimental setup for tracking the deformation is shown in
three steps: (i) tendon-driven process, (ii) pump squeeze, and
Figure 4A.
(iii) pneumatic actuator expansion. In the first step, the tendons
In addition, the blocking force was measured in full operation
are pulled by the motor using a rotating motion (Figure 2B).
with soft finger module, a load cell (GS0 - 500, Transducer
The second step involves compressing the origami pump. The
Techniques, Inc.) is installed on the tip of the pneumatic actuator
ends of the tendons are tied to the bottom part under the soft
with the contact bar. We measure the blocking force in the
origami pump, and they squeeze the pump by winding the tendon
maximum displacement of the tendon-driven origami pump
with the motor. Simultaneously, the air in the origami pump
from the motor operation.
moves into the pneumatic actuator (Figure 2C). The actuation
Moreover, a pressure sensor (TST-10, Nuritech, Inc.) and laser
step is shown in Figure 2D. Specifically, when the air goes to the
sensor (IL-100, KEYENCE Corp.) are utilized to estimate the
soft pneumatic actuator, it pushes the finger structure (Ilievski
pressure output in conjunction with the origami pump operation.
et al., 2011; Deimel and Brock, 2016; Galloway et al., 2016).
A data acquisition board (USB-9162, National Instrument Corp.)
The posterior finger part is designed such that it is stretched
is utilized to measure the outputs from the sensors. The
when the volume of the soft pneumatic actuator is expanded.
experimental setup to observe the pressure from the origami
When the posterior finger part swells, the bending angle of the
pump is shown in Figure 4B.
actuator increases.
In each setups, the motor is operated by the microcontroller
Conversely, the finger release procedure begins with the
(Open CM 9.04, ROBOTIS Co.,Ltd) and internal code to push
tendon-release process. When the motor is driven to the original
and pull tendons.
position, the tension on the tendon is reduced (Figure 2E).
At this time, the origami pump is recovered by the structural
flexibility. This decreases the internal pressure in the soft 3. RESULTS
pneumatic actuator by moving air from the actuator to origami
pump (Figure 2F). Simultaneously, the chamber of the soft 3.1. Finger Operation
pneumatic actuator is reduced, and it ultimately returns to the To characterize the deformation of the soft pneumatic actuator,
initial position (Figure 2G). we analyze the tracking results of the trace markers on the
actuator. The experiment is conducted with a frequency of 0.25
2.3. Experimental Setup Hz and the origami pump height variation (1ĥ) of 13.2 mm
We constructed several experimental setups to evaluate the with the origami pump. Figures 5A,B display the deformation
performance of the soft pneumatic actuator with the tendon- for the bending and release of the pneumatic actuator by sine
driven origami pump. To measure the deformation of the soft input of the tendon-driven origami pump operation, respectively.
pneumatic actuator and the height variation in the origami In addition, Figure 5C shows the full displacements of the tip and
pump, we attach markers to the finger module (Figure 3) and the height variation 1h of the origami pump to compare it with

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 4 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

respectively, at the tip of the pneumatic actuator. Additionally,

we normalize 1x and 1y with the length of the actuator L.
In Figure 5C, we observe 1h is close to the sinusoidal
waveform, while the waveform of 1x presents the distorted wave.
In particular, we observe that the initial and peak regions exhibit
other harmonic terms. This indicates that the observation is due
to the beam oscillation. Additionally, we can compare 0 ∼ 2 s
and 2 ∼ 4 s with bending and releasing sections corresponding
to Figures 5A,B, respectively, through 1h. Within, we observe a
large change in periods corresponding to 2 / T ∼ 4 / T and 6 / T
∼ 8 / T. Furthermore, the results of Figures 5A,B indicate that
the deflection is similar to the fundamental mode beam shape
(Meirovitch and Parker, 2001).

3.2. Height and Pressure Relation

When compared to other pneumatic actuators using pressure or
vacuum pumps (Jittungboonya and Maneewarn, 2019; Zhu et al.,
2019), the amount of working airflow for the pneumatic actuator
is limited by the tendon-driven origami pump. The working
airflow is related to the volume change of the origami pump,
meaning it depends on the 1h of the pump.
Figure 6 shows the demonstration result, and we observe
the hysteresis between the pressure output (p) and 1h. The
experiment is performed at the frequency of 0.25 Hz and the
height variation of 13.2 mm. When the height of the pump is
increased to 8 mm from the original position, the pressure does
not increase. The pressure increases gradually in the range of 8–
12 mm, and increases significantly over 12 mm in compression.
When the height input decreases, the pressure also gradually
decreases. In the large hysteresis, we note that experiments are
FIGURE 5 | Displacement characteristics of pneumatic actuator at f = 0.25
Hz and 1ĥ = 13.2 mm. (A) Normalized 1x and 1y in compression motion of conducted only with the tendon-driven origami pump without a
origami pump. (B) Normalized 1x and 1y in release motion of origami pump. soft pneumatic actuator. The hysteresis can be attributed to the
(C) 1h of origami pump and 1x of actuator for one cycle operation. effects of the fluid dynamics in the origami pump and tendon-
driven operation. In addition, the elasticity of the soft origami
structure can be also one reason for the hysteresis (Chen et al.,
2020).

3.3. Frequency Response

Another important point in the performance of the pneumatic
actuator corresponds to the effect of input actuation frequency
from the tendon-driven origami pump. We conducted
an experiment using the pneumatic actuator at various
frequency inputs (Supplementary Video 1).
Figure 7 presents the result for the various input frequencies
corresponding to 0.6, 1.0, 1.6, 2.3, 2.8, and 3.1 Hz at the
same height variation of 1ĥ = 4.2 mm. It is noted that the
motor exhibits an operational limitation at higher frequencies
exceeding 3.1 Hz in our test setup. Figure 7 shows the
maximum displacement of the x-axis displacement (1x̂) and
phase difference (φ) with the motor input. Each displacement
increases slightly when the input frequency increases, as shown
FIGURE 6 | Pressure output relative to height change of origami pump at f = in Figure 7A. Interestingly, in the vicinity of 3 Hz, it exhibits
0.25 Hz and 1ĥ = 13.2 mm.
significantly increased displacement. Additionally, we show the
phase difference between the displacement of the pneumatic
the input for a period. The positions of the markers are presented actuator and height variation in the origami pump in Figure 7B.
with a time step of 1/T = 0.4 s in Figures 5A,B. Thus, 1x and Also, the phase difference significantly increases when it is closer
1y denote the displacement variations in the x- and y-directions, to 3 Hz. The resonant frequency can be estimated when the height

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 5 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

actuators (Park et al., 2020), and SMA (Lee et al., 2019). Therein,
the soft pneumatic actuators using an external pump produce a
blocking force of up to 1.6767 N with a material with an elastic
modulus of 262.4 kPa and up to 0.5543 N with an elastic modulus
of 48 kPa. It is inferred that the pneumatic actuators using an
external pump can produce a bigger blocking force. In addition,
the blocking forces of the electrohydraulic and SMA actuators are
0.08 N and 0.89 N, respectively. Typically, the blocking force is
proportional to the flexural rigidity of the actuator (Alici et al.,
2018).

3.5. Pressure and Angle Relation

We investigate the degree of curvature of the actuator bending
relative to pressure from the tendon-driven origami pump.
Specifically, we conduct two tests: (i) actuation as the varied
height of the origami pump and (ii) pressure from the pump as
the varied height. The two tests are performed for the same 1ĥ
which corresponds to the 3.2, 4.2, 7.4, 9.3, and 15.3 mm.
Figure 8A shows the relation between normalized 1x̂ and 1ĥ.
The result of 1x̂ is normalized by L, and 1ĥ is normalized by
H. The output result of 1x̂/L tends to proportionally increase
when input height (1ĥ/H) increases. Figure 8B shows the
output pressure relative to the normalized 1ĥ. We also observe
that the pressure from the origami pump increases when the
normalized 1ĥ increases. Thus, we combine each result of 1x̂/L
and p. Specifically, we evaluate the relation between the actuator’s
bending angle (θ ) and p. The bending angle of θ is obtained as
follows (Alici et al., 2018):

1x̂ = R(1 − cosθ )

(1)
L − 1ŷ = Rsinθ
FIGURE 7 | Frequency response of the pneumatic actuator at 1ĥ = 4.2mm.
(A) Maximum displacement at each input frequency. (B) Phase difference at
each input frequency.
where R denotes the radius of the curvature of the bending angle
as shown in Figure 3. Figure 9 and Table 1 show the bending
angle at each input pressure. The bending angle of the pneumatic
actuator relative to pressure was reported in (Alici et al., 2018)
variation (1ĥ) of the origami pump and the x-axis displacement as follows:
of the fingertip (1x̂) have the phase difference of 90◦ (Cao
θ (p) = αp2 + βp (2)
et al., 2019). Moreover, we measure the natural frequency (fn )
to validate the frequency response trend using the free vibration where α = LA2 e
and β = LAe
Aw E 2 I EI , A denotes the inner surface
test from step displacement input (Alici et al., 2008; Aureli
area of the chamber, e denotes the offset between the center of
et al., 2009). Specifically, tweezers are utilized to provide input
pressure and the neutral axis, Aw denotes cross-section area of
displacement, and a camera is used record at a high-speed mode
outside, E denotes the effective elastic modulus of the actuator,
of 120 fps. We measure the natural frequency corresponding to
and I denotes the moment of inertia of the actuator. By fitting
fn = 2.95 Hz from the free vibration test. The value of the natural
our result into Equation (2), we obtain θ (p) = 1.763 × 10−10 p2 +
frequency corresponds to the result shown in Figure 7.
4.829 × 10−5 p [rad] with an R-square value corresponding to
0.9595. The blue solid line in Figure 9 represents the fitting result.
3.4. Blocking Force Furthermore, we can estimate the elastic modulus by using the
The blocking force represents the amount of force provided by coefficients of the fitted equation. In particular, by dividing α by
the pneumatic actuator in a stationary condition (Spinks et al., β, we obtain the following relation
2005). This makes it possible to infer the operating characteristics
of the actuator and to estimate the weight of the object that the α A
gripper can handle. We obtain the blocking force of 0.0653 N in = (3)
β Aw E
the full operation. Within, we compare the blocking force of the
actuator with other actuators such as pneumatic actuators with where A and Aw are obtained from the 3D CAD program, and the
an external pressure pump (Alici et al., 2018), electrohydraulic values are 2.07 × 10−4 mm2 and 5.72 × 10−5 mm2 , respectively.

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 6 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

FIGURE 9 | Measurement results and fitted data of actuator’s bending angle

and pressure from the origami pump.

TABLE 1 | Bending angle of pneumatic actuator at different pressure values.

p [kPa] θ [deg]

4.25 22.08
9.25 26.38
16.67 49.52
24.18 64.25
33.56 107.95

Moreover, the natural frequency of the pneumatic actuator is

as follows (Meirovitch and Parker, 2001):
FIGURE 8 | (A) Relation between normalized pump height and actuator s
displacement in x-direction. (B) Relation between normalized height and 3.5160 EI
output pressure of the origami pump. Test input frequency corresponds to fn = (5)
2π ml L4
f = 1 Hz.

where ml = 0.11 kg/m is mass per unit length. By applying the

elastic modulus of 43.28 kPa, the moment of inertia is estimated
In this case, the elastic modulus is calculated as 990 kPa. The as 1.80 × 10−9 kg· m2 .
calculated value is relatively larger than the elastic modulus of
the actuator material, which is known as 30-43 kPa (Larson et al., 3.6. Soft Gripper
2016; Yang et al., 2016; Jang et al., 2017). Also, we observe that We designed a soft gripper using the three proposed finger
the first test point does not match the fitted line. We assume that modules with each motor. Each module is positioned at each
the actuator exhibits an initial curve, and we attempt to modify angular point of the triangle, as shown in Figure 10A. The
the relation as follows: body of the gripper consists of an acrylic plate, which is
manufactured with a laser cutter. The completed soft pneumatic
gripper system with tendon-driven origami pump is shown in
θ (p) = αp2 + βp + γ (4) Figure 10B, and the grasping motion of the gripper is shown in
Supplementary Video 2.
where γ denotes the constant term by the initial curve of the Within, we measured the displacements of the three finger
pneumatic actuator. The red dashed line in Figure 9 shows the modules with the experimental setup shown in Figure 4A.
fitted data given the initial curve. The fitting result corresponds Figure 11A describes the initial position of each pneumatic
to θ (p) = 1.017 × 10−9 p2 + 1.217 × 10−5 p + 0.3089 [rad] with actuator module, and Figures 11B,C present the maximum
R-square value of 0.9877. We estimate that the actuator exhibits displacements at 1ĥ/H = 0.13 and 0.19, respectively. The trend
the initial curve corresponding to 17 degrees. Additionally, we of initial position and maximum displacement of each pneumatic
use the relation between α and β and obtain E = 43.28 kPa, actuator are morphologically similar at same inputs. However,
which is compatible with that in previous studies (Larson et al., we can observe the deviation of each displacement caused by
2016; Yang et al., 2016; Jang et al., 2017). the handmade fabrication procedure. This results in a height

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 7 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

FIGURE 10 | (A) Design of soft gripper with three soft finger modules. (B) Photograph of full soft gripper system.

FIGURE 11 | Position of the pneumatic actuator relative to the height variation of the origami pump. (A) 1ĥ/H = 0, (B) 1ĥ/H = 0.13, and (C) 1ĥ/H = 0.19.

difference of 2.5 mm between the tip of the actuators during shows the feasibility of the pneumatic gripper without external
the operation. air compressor.
Moreover, we conducted several grasping experiments with
diverse materials and shapes. The capability of the soft
gripper performance is shown in Figure 12. Specifically, the 4. CONCLUSIONS
soft gripper grasps an aluminum cup, balloon, table tennis
ball, and snack (Supplementary Video 3). The results indicate In this study, we designed a soft pneumatic actuator with
that the soft gripper exhibits novelty in terms of grasping a tendon-driven soft origami pump. We conducted a series
brittle and crumbly objects. Additionally, the demonstration of experiments and analyses to evaluate the performance

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 8 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

FIGURE 12 | Grasping performance of soft gripper with (A) an aluminum cup, (B) a balloon (C) a table tennis ball, and (D) a ring shaped snack.

of the actuator. Specifically, we examined the movement DATA AVAILABILITY STATEMENT

characteristics, frequency response, blocking force, and
pressure and bending angle relation. In addition, we All datasets generated for this study are included in the
demonstrated the grasping of the soft gripper consisting of three article/Supplementary Material.
pneumatic actuators.
The results indicated that the operation of the pneumatic
AUTHOR CONTRIBUTIONS
actuator exhibits the hysteresis characteristics. We suspected that YK and YC planned and designed the experiments. YK
the characteristic is caused by the fluid dynamic effect of the performed the fabrication and experiment. YK and YC analyzed
origami pump, elasticity of the origami structure, tendon-driven the data and wrote the manuscript.
operation, and soft pneumatic actuator. Additionally, the results
of the frequency response experiment suggested that the natural FUNDING
frequency of the pneumatic actuator is approximately 3 Hz.
Furthermore, we derived the relation between the bending angle This work was supported by the KIST flagship program under
of the pneumatic actuator and pressure from the origami pump. Project 2E30280.
The bending angle corresponded the second order polynomial
function of the pressure. Also, we calculated effective elastic ACKNOWLEDGMENTS
modulus of the actuator from the experimental data, and the
The authors would like to thanks Chohee Kim for her help with
value was comparable to that in extant studies.
drawing schematic.
From the practical and methodological perspectives, the main
contributions of our studies are as follows: 1) addressing the SUPPLEMENTARY MATERIAL
feasibility of a soft origami pump using the tendon-driven
principle with a pneumatic actuator, and 2) demonstrating a The Supplementary Material for this article can be found
soft pneumatic gripper module that can be utilized as a robot online at: https://www.frontiersin.org/articles/10.3389/fbioe.
manipulator without external air supplying equipment. 2020.00461/full#supplementary-material

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 9 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

REFERENCES Kobayashi, H., Kresling, B., and Vincent, J. F. (1998). The geometry of
unfolding tree leaves. Proc. R. Soc. Lond. Ser. B Biol. Sci. 265, 147–154.
Alici, G., Canty, T., Mutlu, R., Hu, W., and Sencadas, V. (2018). Modeling doi: 10.1098/rspb.1998.0276
and experimental evaluation of bending behavior of soft pneumatic Kofod, G., Wirges, W., Paajanen, M., and Bauer, S. (2007). Energy minimization for
actuators made of discrete actuation chambers. Soft Robot. 5, 24–35. self-organized structure formation and actuation. Appl. Phys. Lett. 90:081916.
doi: 10.1089/soro.2016.0052 doi: 10.1063/1.2695785
Alici, G., Spinks, G. M., Madden, J. D., Wu, Y., and Wallace, G. G. (2008). Response Kresling, B. (2008). “Natural twist buckling in shells: From the Hawkmoth’s bellows
characterization of electroactive polymers as mechanical sensors. IEEE/ASME to the deployable Kresling-pattern and cylindrical Miuraori,” in Proceedings of
Trans. Mechatron. 13, 187–196. doi: 10.1109/TMECH.2008.918531 the 6th International Conference on Computation of Shell and Spatial Structures,
Aureli, M., Prince, C., Porfiri, M., and Peterson, S. D. (2009). Energy harvesting eds J. F. Abel and J. R. Cooke (Ithaca, NY), 12–32.
from base excitation of ionic polymer metal composites in fluid environments. Lang, R. J. (2011). Origami Design Secrets: Mathematical Methods for an Ancient
Smart Mater. Struct. 19:015003. doi: 10.1088/0964-1726/19/1/015003 Art. Natick: AK Peters/CRC Press. doi: 10.1201/b11074
Bhattacharya, S., Bepari, B., and Bhaumik, S. (2019). “Design and fabrication Larson, C. E., Peele, B. N., Li, S., Robinson, S. A., Totaro, M., Beccai, L.,
of deformable soft gripper using IPMC as actuator,” in Ionic Polymer Mazzolai, B., and Shepherd, R. (2016). Highly stretchable electroluminescent
Metal Composites for Sensors and Actuators (Berlin: Springer), 195–207. skin for optical signaling and tactile sensing. Science 351, 1071–1074.
doi: 10.1007/978-3-030-13728-1_10 doi: 10.1126/science.aac5082
Butterfaß, J., Grebenstein, M., Liu, H., and Hirzinger, G. (2001). “DLR-hand II: Laschi, C., Mazzolai, B., Mattoli, V., Cianchetti, M., and Dario, P. (2009).
Next generation of a dextrous robot hand,” in Proceedings 2001 ICRA. IEEE Design of a biomimetic robotic octopus ARM. Bioinspir. Biomimet. 4:015006.
International Conference on Robotics and Automation (Cat. No. 01CH37164), doi: 10.1088/1748-3182/4/1/015006
Vol. 1 (Seoul: IEEE), 109–114. doi: 10.1109/ROBOT.2001.932538 Lee, J.-G., and Rodrigue, H. (2019). Origami-based vacuum pneumatic
Cao, C., Burgess, S., and Conn, A. T. (2019). Toward a dielectric elastomer artificial muscles with large contraction ratios. Soft Robot. 6, 109–117.
resonator driven flapping wing micro air vehicle. Front. Robot. AI 5:137. doi: 10.1089/soro.2018.0063
doi: 10.3389/frobt.2018.00137 Lee, J.-H., Chung, Y. S., and Rodrigue, H. (2019). Long shape memory alloy
Carrico, J. D., and Leang, K. K. (2017). “Fused filament 3D printing of ionic tendon-based soft robotic actuators and implementation as a soft gripper. Sci.
polymer-metal composites for soft robotics,” in Electroactive Polymer Actuators Rep. 9, 1–12. doi: 10.1038/s41598-019-47794-1
and Devices (EAPAD) 2017, Vol. 10163 (Portland: International Society for Li, H., Go, G., Ko, S. Y., Park, J.-O., and Park, S. (2016). Magnetic actuated ph-
Optics and Photonics), 101630I. doi: 10.1117/12.2259782 responsive hydrogel-based soft micro-robot for targeted drug delivery. Smart
Chan, V., Park, K., Collens, M. B., Kong, H., Saif, T. A., and Bashir, R. (2012). Mater. Struct. 25:027001. doi: 10.1088/0964-1726/25/2/027001
Development of miniaturized walking biological machines. Sci. Rep. 2:857. Li, S., Vogt, D. M., Rus, D., and Wood, R. J. (2017). Fluid-driven origami-
doi: 10.1038/srep00857 inspired artificial muscles. Proc. Natl. Acad. Sci. U.S.A. 114, 13132–13137.
Chen, W.-H., Misra, S., Gao, Y., Lee, Y.-J., Koditschek, D. E., Yang, S., et al. (2020). doi: 10.1073/pnas.1713450114
A programmably compliant origami mechanism for dynamically dexterous Lu, X., Xu, W., and Li, X. (2017). A soft robotic tongue-mechatronic design
robots. IEEE Robot. Autom. Lett. 5, 2131–2137. doi: 10.1109/LRA.2020.2970637 and surface reconstruction. IEEE/ASME Trans. Mechatron. 22, 2102–2110.
Deimel, R., and Brock, O. (2016). A novel type of compliant and underactuated doi: 10.1109/TMECH.2017.2748606
robotic hand for dexterous grasping. Int. J. Robot. Res. 35, 161–185. Mahadevan, L., and Rica, S. (2005). Self-organized origami. Science 307,
doi: 10.1177/0278364915592961 1740–1740. doi: 10.1126/science.1105169
Della Santina, C., Arapi, V., Averta, G., Damiani, F., Fiore, G., Settimi, A., et al. Marchese, A. D., Onal, C. D., and Rus, D. (2014). Autonomous soft robotic fish
(2019). Learning from humans how to grasp: a data-driven architecture for capable of escape maneuvers using fluidic elastomer actuators. Soft Robot. 1,
autonomous grasping with anthropomorphic soft hands. IEEE Robot. Autom. 75–87. doi: 10.1089/soro.2013.0009
Lett. 4, 1533–1540. doi: 10.1109/LRA.2019.2896485 Meirovitch, L., and Parker, R. G. (2001). Fundamentals of vibrations. Appl. Mech.
Demaine, E. D., and O’Rourke, J. (2007). Geometric Folding Algorithms: Rev. 54, B100–B101. doi: 10.1115/1.1421112
Linkages, Origami, Polyhedra. Cambridge: Cambridge University Press. Mitchell, S. K., Wang, X., Acome, E., Martin, T., Ly, K., Kellaris, N.,
doi: 10.1017/CBO9780511735172 et al. (2019). An easy-to-implement toolkit to create versatile and high-
Firouzeh, A., and Paik, J. (2017). An under-actuated origami gripper with performance Hasel actuators for untethered soft robots. Adv. Sci. 6:1900178.
adjustable stiffness joints for multiple grasp modes. Smart Mater. Struct. doi: 10.1002/advs.201900178
26:055035. doi: 10.1088/1361-665X/aa67fd Onal, C. D., Wood, R. J., and Rus, D. (2012). An origami-inspired
Galloway, K. C., Becker, K. P., Phillips, B., Kirby, J., Licht, S., Tchernov, D., et al. approach to worm robots. IEEE/ASME Trans. Mechatron. 18, 430–438.
(2016). Soft robotic grippers for biological sampling on deep reefs. Soft Robot. doi: 10.1109/TMECH.2012.2210239
3, 23–33. doi: 10.1089/soro.2015.0019 Paez, L., Agarwal, G., and Paik, J. (2016). Design and analysis of a soft
Hong, M. B., Kim, S. J., Ihn, Y. S., Jeong, G.-C., and Kim, K. (2018). Kulex-hand: pneumatic actuator with origami shell reinforcement. Soft Robot. 3, 109–119.
An underactuated wearable hand for grasping power assistance. IEEE Trans. doi: 10.1089/soro.2016.0023
Robot. 35, 420–432. doi: 10.1109/TRO.2018.2880121 Palagi, S., Mark, A. G., Reigh, S. Y., Melde, K., Qiu, T., Zeng, H., et al. (2016).
Hughes, J., Culha, U., Giardina, F., Guenther, F., Rosendo, A., and Iida, F. Structured light enables biomimetic swimming and versatile locomotion of
(2016). Soft manipulators and grippers: a review. Front. Robot. AI, 3:69. photoresponsive soft microrobots. Nat. Mater. 15:647. doi: 10.1038/nmat4569
doi: 10.3389/frobt.2016.00069 Park, T., Kim, K., Oh, S.-R., and Cha, Y. (2020). Electrohydraulic actuator for a soft
Ilievski, F., Mazzeo, A. D., Shepherd, R. F., Chen, X., and Whitesides, G. M. gripper. Soft Robot. 7, 68–75. doi: 10.1089/soro.2019.0009
(2011). Soft robotics for chemists. Angew/ Chem. Int. Ed. 50, 1890–1895. Rodrigue, H., Wang, W., Kim, D.-R., and Ahn, S.-H. (2017). Curved shape memory
doi: 10.1002/anie.201006464 alloy-based soft actuators and application to soft gripper. Composite Struct. 176,
Jang, K.-I., Li, K., Chung, H. U., Xu, S., Jung, H. N., Yang, Y., et al. (2017). 398–406. doi: 10.1016/j.compstruct.2017.05.056
Self-assembled three dimensional network designs for soft electronics. Nat. Roy, S., Kim, J., Kotal, M., Tabassian, R., Kim, K. J., and Oh, I.-K. (2019).
Commun. 8:15894. doi: 10.1038/ncomms15894 Collectively exhaustive electrodes based on covalent organic framework and
Jianguo, C., Xiaowei, D., Ya, Z., Jian, F., and Yongming, T. (2015). Bistable behavior antagonistic co-doping for electroactive ionic artificial muscles. Adv. Funct.
of the cylindrical origami structure with kresling pattern. J. Mech. Design Mater. 29:1900161. doi: 10.1002/adfm.201900161
137:061406. doi: 10.1115/1.4030158 Salerno, M., Zhang, K., Menciassi, A., and Dai, J. S. (2016). A novel 4-dof origami
Jittungboonya, P., and Maneewarn, T. (2019). “Grasping with a tube-feet inspired grasper with an SMA-actuation system for minimally invasive surgery. IEEE
soft gripper,” in 2019 First International Symposium on Instrumentation, Trans. Robot. 32, 484–498. doi: 10.1109/TRO.2016.2539373
Control, Artificial Intelligence, and Robotics (ICA-SYMP) (Bangkok: IEEE), Snell-Rood, E. (2016). Interdisciplinarity: bring biologists into biomimetics. Nat.
147–150. doi: 10.1109/ICA-SYMP.2019.8646007 News 529:277. doi: 10.1038/529277a

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 10 May 2020 | Volume 8 | Article 461
Kim and Cha Pneumatic Gripper With Origami Pump

Spinks, G. M., Campbell, T. E., and Wallace, G. G. (2005). Force generation from hand controlled by pulse modulation. Mechatronics 15, 589–609.
polypyrrole actuators. Smart Mater. Struct. 14:406. doi: 10.1088/0964-1726/14/ doi: 10.1016/j.mechatronics.2004.09.007
2/015 Yang, D., Verma, M. S., So, J.-H., Mosadegh, B., Keplinger, C., Lee, B., et al. (2016).
Sun, Y., Song, Y. S., and Paik, J. (2013). “Characterization of silicone Buckling pneumatic linear actuators inspired by muscle. Adv. Mater. Technol.
rubber based soft pneumatic actuators,” in 2013 IEEE/RSJ International 1:1600055. doi: 10.1002/admt.201600055
Conference on Intelligent Robots and Systems (Tokyo: IEEE), 4446–4453. Zhu, M., Mori, Y., Wakayama, T., Wada, A., and Kawamura, S. (2019). A fully
doi: 10.1109/IROS.2013.6696995 multi-material three-dimensional printed soft gripper with variable stiffness for
Taghavi, M., Helps, T., and Rossiter, J. (2018). Electro-ribbon actuators and electro- robust grasping. Soft Robot. 6, 507–519. doi: 10.1089/soro.2018.0112
origami robots. Sci. Robot. 3:eaau9795. doi: 10.1126/scirobotics.aau9795
Tolley, M. T., Shepherd, R. F., Mosadegh, B., Galloway, K. C., Conflict of Interest: The authors declare that the research was conducted in the
Wehner, M., Karpelson, M., et al. (2014). A resilient, untethered absence of any commercial or financial relationships that could be construed as a
soft robot. Soft Robot. 1, 213–223. doi: 10.1089/soro.2014. potential conflict of interest.
0008
Wang, W., Lee, J.-Y., Rodrigue, H., Song, S.-H., Chu, W.-S., and Ahn, S.- Copyright © 2020 Kim and Cha. This is an open-access article distributed under the
H. (2014). Locomotion of inchworm-inspired robot made of smart soft terms of the Creative Commons Attribution License (CC BY). The use, distribution
composite (SSC). Bioinspir. Biomimet. 9:046006. doi: 10.1088/1748-3182/9/4/ or reproduction in other forums is permitted, provided the original author(s) and
046006 the copyright owner(s) are credited and that the original publication in this journal
Wojtara, T., Nonami, K., Shao, H., Yuasa, R., Amano, S., Waterman, is cited, in accordance with accepted academic practice. No use, distribution or
D., et al. (2005). Hydraulic master-slave land mine clearance robot reproduction is permitted which does not comply with these terms.

Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 11 May 2020 | Volume 8 | Article 461