ArticlePDF Available

Soft Fingertips With Tactile Sensing and Active Deformation for Robust Grasping of Delicate Objects

Authors:

Abstract and Figures

Soft fingertips have shown significant adaptability for grasping a wide range of object shapes thanks to elasticity. This ability can be enhanced to grasp soft, delicate objects by adding touch sensing. However, in these cases, the complete restraint and robustness of the grasps have proved to be challenging, as the exertion of additional forces on the fragile object can result in damage. This paper presents a novel soft fingertip design for delicate objects based on the concept of embedded air cavities, which allow the dual ability of adaptive sensing and active shape changing. The pressurized air cavities act as soft tactile sensors to control gripper position from internal pressure variation; and active fingertip deformation is achieved by applying positive pressure to these cavities, which then enable a delicate object to be kept securely in position, despite externally applied forces, by form closure. We demonstrate this improved grasping capability by comparing the displacement of grasped delicate objects exposed to high-speed motions. Results show that passive soft fingertips fail to restrain fragile objects at accelerations as low as ${0.1 m/s^{2}}$ , in contrast, with the proposed fingertips, delicate objects are completely secure even at accelerations of more than ${5 m/s^{2}}$ .
Content may be subject to copyright.
IEEE ROBOTICS AND AUTOMATION LETTERS. PREPRINT VERSION. ACCEPTED JANUARY, 2020 1
Soft Fingertips with Tactile Sensing and
Active Deformation for Robust Grasping of
Delicate Objects
Liang He1, Qiujie Lu1, Sara-Adela Abad2,3, Nicolas Rojas1, and Thrishantha Nanayakkara1
Abstract—Soft fingertips have shown significant adaptability
for grasping a wide range of object shapes thanks to elasticity.
This ability can be enhanced to grasp soft, delicate objects by
adding touch sensing. However, in these cases, the complete
restraint and robustness of the grasps have proved to be challeng-
ing, as the exertion of additional forces on the fragile object can
result in damage. This paper presents a novel soft fingertip design
for delicate objects based on the concept of embedded air cavities,
which allow the dual ability of tactile sensing and active shape-
changing. The pressurized air cavities act as soft tactile sensors
to control gripper position from internal pressure variation; and
active fingertip deformation is achieved by applying positive
pressure to these cavities, which then enable a delicate object to
be kept securely in position, despite externally applied forces, by
form closure. We demonstrate this improved grasping capability
by comparing the displacement of grasped delicate objects
exposed to high-speed motions. Results show that passive soft
fingertips fail to restrain fragile objects at accelerations as low
as 0.1m/s2, in contrast, with the proposed fingertips delicate
objects are completely secure even at accelerations of more than
5m/s2.
Index Terms—Soft Sensors and Actuators, Grasping, Gripper
and Other End-Effectors.
I. INTRODUCTION
The need for robotic grasping and manipulating soft objects
without damage has been growing in farming, industrial, and
household environments [1]. Soft and fragile objects such as
fruits, vegetables, or other non-rigid food products tend to
either break or deform if too much force is applied [2]. Both
the agricultural sector and the food industry are currently
urging for such grippers in the continuous automation of
their processes, for increased efficiency, hygiene, and quality—
e.g., [3], Human workers can easily handle delicate products
without crushing them for tasks such as gripping, positioning,
orienting, and placing. In contrast, robot grippers face great
Manuscript received: September, 10, 2019; Revised November, 17, 2019;
Accepted January, 20, 2020.
This paper was recommended for publication by Editor Kyu-Jin Cho upon
evaluation of the Associate Editor and Reviewers’ comments. *This work was
supported in part by the Engineering and Physical Sciences Research Council
grant EP/R020833/1.
1Liang He, Qiujie Lu, Nicolas Rojas, and Thrishantha Nanayakkara are
with the Dyson School of Design Engineering, Imperial College London, 25
Exhibition Road, London SW7 2DB, UK. L.he17@imperial.ac.uk
2Sara-Adela Abad is with the Department of Mechanical Engineering, Uni-
versity College London, London, UK s.abad-guaman@ucl.ac.uk
3Sara-Adela Abad is also with the Institute for Applied Sustainability
Research, Av. Granados E13-55 e Isla Marchena, No.44, Quito, 170503,
Ecuador.
Digital Object Identifier (DOI): see top of this page.
Fig. 1. Embedded air cavities in the proposed soft fingertips allow the
dual ability of soft tactile sensing, via internal pressure variation, and active
deformation, via positive pressure. The active shape-changing allows keeping
delicate objects securely in position despite external forces being applied.
difficulties to grasp soft and fragile objects with different size,
stiffness, and shape [4]. Natural food products like fruits, for
instance, tend to get bruised when high contact pressure is
applied on the contact point [5]. In addition, the modeling
of the contact point and the gripper is limited due to the
shape and ripeness variation of natural fruits. The grasping also
needs to be robust enough to be manipulated in high speed for
production efficiency. The challenge is how to ensure precision
robust grasping with controllable applied force on soft objects.
Currently, rigid contact grippers [6], soft robotic grippers
[7], and air with suction cups [8] are designed to be used
in the food industry for pick and place tasks. Rigid grippers
have the advantages of high precision control with the ease of
integrating tactile, position, and torque sensors [9]. A stable
grasping force and dexterous manipulation of the object can be
achieved with force and position feedback [10]. However, rigid
grippers are lacking adaptability, and they are prone to cause
damages when manipulating very fragile and soft objects.
Conversely, soft grippers such as the starfish gripper with
the embedded pneumatic networks (PneuNets) [11] and phase
changing grippers like the universal gripper with granular
jamming material [12] show high resilience and compliance.
The high adaptability of soft robotic grippers indeed increased
the performance in grasping delicate food products with lower
contact pressure, yet the precise modeling and robustness are
very challenging compared to rigid grippers [13]. In addition,
most grasping tasks with soft grippers are enabled by power
grasp, which disabled the capability for in-hand and precision
2 IEEE ROBOTICS AND AUTOMATION LETTERS. PREPRINT VERSION. ACCEPTED JANUARY, 2020
manipulation [14]. Specialized grasping technologies based on
airflow such as suction cups and Bernoulli principle grippers
are possible to handle objects with different sizes and work
with multiple objects at the same time [15], [8]. However,
the airflow-based technologies work poorly with products that
have rough surfaces, porous structure, and irregular shapes.
The product surface can also get damaged or marked during
grasping and manipulation.
One solution to achieve robust soft object grasping is to
enable form closure and tactile sensing with fingertips that
can change the morphology. Form closure is created when
the object is fully constrained by contacts of the grippers,
irrespective of the magnitude of the forces [16], [17]. Thus,
the required force to prevent the object from escaping can
be significantly reduced in industrial pick and placing tasks
compared to two-point pitch grasp [18] which relies on friction
(force closure). The increased contact areas between the object
and gripper can also improve the grasping stability as shown
in [19], where the shape of the fingertips are designed to adapt
the object based on contact primitives. Tactile sensors, such
as capacitive [20], optical [10], and piezoelectric [21] sensors
are commonly being employed on the fingers for sensing
to detect grasping force and increase accuracy [22], [10].
Fingertip designs with embedded sensors can also achieve high
resolution and relatively soft contact when interacting with
the object [23], [24]. However, conventional tactile sensors
are difficult to integrate with high softness grippers and soft
fingertips that can produce active deformation.
In this paper, we propose a novel fingertip design that has
the dual ability of tactile sensing and active shape-changing
(Fig. 1) with the same set of pneumatic-controlled hardware.
The fingertips are integrated and tested with a traditional
grasper with four-bar-linkage fingers based on the GR2 gripper
[25]. The grasper itself without the designed fingertips is
inappropriate for soft object grasping as a high contact force is
needed for securing the object in hand. Our aim is to combine
the advantages of soft robot design with a rigid gripper to
change the soft fingertip geometry via positive pressures and
use pressure variations for tactile sensing. By active changing
the shape of the fingertips to adapt to the object shape,
implementations such as 2D form-closure grasping can be
created in regard to the object sizes. Combining the dual
function of the soft fingertips allows the gripper to be able to
grasp soft and delicate objects with low contact force, while
being robust enough for high-speed manipulations.
The rest of the paper is organized as follows. Section II
introduces the design and control of the dual-soft fingertips
and modified gripper. Experiment results of crushing tests with
sensorised soft cylindrical objects of different size and escap-
ing tests with these at high speed are detailed in section III
and IV, respectively. Finally, section V provides a conclusion
and discussion of the work.
II. FI NG ERT IP DE SI GN A ND CO NT ROL
A. Fingertip design
The fingertip structure was designed with the consideration
of the planar grasping motion for demonstrating the proposed
Fig. 2. (a) Schematics of the dual-function fingertip design. (b) Fingertip
thickness wand (c) shore hardness variation of inflation (blue) and deflation
(red) tests. The x-axis label of b is the same as c. (d) Sensor output values
versus applied surface load for fingertips of different inflated pressure.Three
lines per inflated condition were shown with loading and unloading test cycles.
dual-functionality. The soft fingertip is designed with silicone
rubber (Ecoflex 0030 with part A, part B mixing ratio of 1:1,
from Smooth-on, USA) to ensure safe interaction with soft
objects. Two half-cylindrical air cavities made from the same
silicone rubber are created and cast inside the soft fingertip
for sensing and inflation. The base of the soft fingertip is 3D
printed with VeroClear on a Stratasys Object 260 3D printer.
The structure of the base is designed to ensure the silicone
fingertip fully secures on the base and grow in the desired
direction during inflation. Fig. 2a. shows the schematics of the
dual-function fingertip design. This figure also demonstrates
a scenario where the fingertips are inflated to adapt to the
object shape actively, and a 2D form-closure grasping can be
therefore achieved. With different inflated air cavity sizes, the
fingertips can be controlled to adapt to the grasping object.
According to the ideal gas law, PV =nRT (where, P,V, and
Tare the pressure, volume, and absolute temperature, n,Rare
the number of moles of gas and the ideal gas constant), the
internal air pressure shows a co-relation to the volume change
of the air cavity when n,R,Tare constant. Considering the
incompressibility of silicone, the volume change of the air
cavity caused by external force can be measured with a pres-
sure sensor. Previous work [26] about pneumatic deformation
HE et al.: SOFT FINGERTIPS WITH TACTILE SENSING AND ACTIVE DEFORMATION 3
Fig. 3. (a) Kinematic chain of the gripper during grasping task. P1, P2, P3, and P4 are revolute joints on the left gripper. P5 is the middle point of the
fingertip rigid surface. P6 is the middle point of the silicone surface of the fingertip. θis the input angle to close/open the mechanism. ρis the fixing angle
of the central triangular link after locked in place. D is the distance between two grippers measured from the middle points. The gripper moves symmetric
during the grasping task in this study. (b) Sample data of the pressure reading of the four air cavites when grasping a 30mm cylinder. The four subplots are
represented from the four sensing air cavities (L1, L2, R1, and R4), respectively.
sensing confirms that the air cavity can be used as a force
sensor with the reading from pressure sensors.
We use the embedded air cavities as soft tactile sensors to
control the gripper position with the internal pressure data.
When the air cavity is inflated, the deformation of the air
cavity causes shape change of the gripper’s fingertip. The
mechanical properties (size and hardness) of the fingertip
were tested and shown in Fig. 2. Thickness wand shore
hardness were collected at the widest part of the fingertip
with a caliper and a type C Durometer (resolution of 0.5 HC),
respectively. For a single trial, the fingertip was inflated by
step around 5 kPa to measure those mechanical properties,
same as the deflation. The curves represent the average data of
5 trials of inflation/deflation tests in a continuous cycle. The
fingertip thickness wincreases from 22 mm to 29 mm and
the shore hardness of the fingertip increases from 20 HC to
33.5 HC during inflation (102 kPa to 150 kPa). The measured
shore hardness of the original Ecoflex 0030 is 25 HC. The
fingertip exhibits a hardness change during inflation from
softer to stiffer compared to original fingertips. Fig. 2 also
shows that the hysteresis is 0.005 for thickness and 0.062
for shore hardness as a consequence of material hysteresis.
The maximum measured variability was 0.9 mm for thickness
and 2.3 HC for shore hardness. The sensitivity of the tactile
sensor was evaluated by applying a load to the center of one of
the fingertip cavities. A 3D printed probe with the tip diameter
of 20 mm was mounted on an ANT130 XY-stage (Aerotech
Inc., accuracy of 2.5 µm) with an ATI Mini40-E Force/Torque
sensor (SI-40-2, ATI Industrial Automation, USA, resolution
of 0.02 N) attached to perform the calibration. The load was
applied incrementally until the sensor reading saturated. Due to
the maximum intake load of the XY-stage, the maximum force
was recorded at around 20 N. A clear increase of sensing range
was observed when inflating the air cavity. The maximum
sensing range of the non-inflated state cavity is around 10 N,
Fig. 4. Pneumatic diagram for the fingertip control. All 4 pressurized
air cavities are controlled and measured independently. The dashed square
indicates the components of the fingertips while pressure sensors, solenoid
valves, the air reservoir, and the air pump are connected externally. Black lines
indicate the pneumatic connections, red lines indicate the sensors signals, and
blue lines indicate control signals.
while the maximum range increased to more than 20 N when
the air cavity is inflated to 50 kPa. The load was also gradually
removed to evaluate the hysteresis: no visible hysteresis was
shown for the inflated pressure 0,10,25,and 40 kPa; small
hysteresis was shown with the 50 kPa. Sensor output with
load showed good linearity (r2>0.99) for the inflated pressure
0,10,25,and 40. Data for inflated pressure 50 kPa showed
lower linearity with average r2=0.98. Signals were recorded
at 1000 Hz. The system showed good sensitivity with less than
0.05 N with characterization method found in [24] (0.0287N,
0.0263N, 0.0222N, 0.0218N, and 0.0351N for the inflated
pressure 0,10,25,40 and 50 kPa). Additional filtering can
further reduce the signal noise to increase the sensitivity and
reliability of the fingertips. Data are shown in Fig. 2d.
B. Gripper and fingertips control
We redesigned the GR2 gripper with a gear mechanism
instead of a tendon-driven actuation to increase the actuation
4 IEEE ROBOTICS AND AUTOMATION LETTERS. PREPRINT VERSION. ACCEPTED JANUARY, 2020
Fig. 5. 3D printed soft cylinders with pneumatic sensors in variation sizes.
accuracy, as well as locked the central triangular link to
simplify the grasping stage. Two Dynamixel MX-64T servo
motors (ROBOTIS, South Korea) were used to control the
gears via a U2D2 USB communication converter with Labview
2018. Fig. 3a. shows the kinematic chain of the gripper during
the grasping task. The linkage between P1 and P4 are fixed in
this study to make the GR2 gripper perform as a traditional
fixed-geometry base two-finger robot gripper. During grasping
tasks, the soft air cavities on the fingertips deform when the
gripper grasps an object. The sensing data in Fig. 3 shows the
demonstrated pressure variations of all four air cavities (L1,
L2 on the left fingertip and R1, R2 on the right fingertip) when
grasping a 30mm cylindrical object.
The pneumatic diagram in Fig. 4 shows the connection
of the four soft air cavities in the fingertips and external
hardware. The air cavities were connected with 4 pressure
sensors (PSE 543-R06, SMC Corporation, Japan), 5 solenoid
valves (Z031C, Sirai, Italy), one air reservoir, and an air pump.
A National Instrument DAQ (USB-6341) with Labview 2018
was used to measure the internal air pressure and control the
experiment. The internal air pressure of each soft air cavities
(L1, L2, R1, R2) was controlled separately with a closed-loop
pressure controller. The pressure increase was achieved by a
proportional controller with the air pump while a computer-
controlled solenoid valve (S5) was used in the system to
decrease the pressure. Since the air cavities have a small
volume, an air reservoir was used to stabilize the internal
pressure without overshoot. When closing the solenoid valves
S1, S2, S3, and S4, the internal air mass in the air cavity was
maintained. Then, the air cavities L1, L2, R1, and R2 can
be used for sensing tasks by measuring the internal pressure
variations through the pressure sensors P1, P2, P3, and P4,
respectively. When the shape-changing of the fingertips is
needed, the controller will be used to inflate the soft air cavities
to desired pressure conditions.
III. CRUS HI NG TEST WITH SO FT OBJECTS
A. Soft testing Objects
The experiment focuses on demonstrating the sensing ca-
pability and repeatability of the tactile sensors during soft
object grasping. In order to quantify the impact of the gripper
on soft objects, 3D printed soft cylinders with the same
concept of pneumatic sensing of the fingertips were used in
this study. The soft cylinders are airtight and connected to
a pressure sensor where the internal pressure change of the
cylinder represents the amount of deformation caused by an
Fig. 6. Type 1 fingertips with only silicone pad and type 2 fingertips with
pressurized air cavities
external force. Five soft sensing cylinders with the diameter
D0=10,20,30,40,50 mm, length of 50 mm, and 1.5 mm
thickness were 3D printed using combinations of TangoPlus
and VeroClear on an Object 260 3D printer [27] with the
material FLX9960-DM (Fig. 5).
B. Gripper position control with tactile sensors
Two types of fingertips (Fig. 6) were tested to show the
comparison results on grasping soft cylindrical objects. Firstly,
only position control of the gripper with fingertip type 1
(pure silicone fingertip with no sensor) were used to grasp
the soft object to the distance between two fingers D=0.9D0,
D=0.8D0,D=0.7D0, and D=0.6D0, where D0is the initial
diameter of the soft object without deformation. The smaller
the distance Dbetween the two fingers, the larger deformation
will be caused by the grasp. The relative pressure change
inside the soft object reflects the deformation quantitatively.
During the test, each object was placed in the center of
the grippers. This procedure was repeated 5 times for each
grasping condition. Secondly, position control with fingertip
tactile sensors was enabled with fingertip type 2 to detect
the object during grasping. We implemented a closed-loop
controller based on the pressure information from the 4 air
cavities on the fingertips to control the input angle θof the
gripper during the test [28]. The controller is described as
θk=
θk1+˙
θdtif P
kP
s
θk1if P
s<P
kP
t
θk1˙
θdtif P
t<P
k
(1)
and
θmin θkθmax (2)
where, θkis the present angle of the motor, P
kis the highest
pressure change among the 4 sensed pressures from L1, L2,
R1, and R2 (to mitigate the grasping position errors), ˙
θdis
the reference velocity to open/close gripper, tis the time
step, P
sand P
tare the lower and higher band pressure (within
sensor sensing range, see Fig. 2d), and θmin and θmax are the
minimum and maximum rotation angle of the gripper. During
the tactile sensing experiment, P
swas set to be 0.1 kPa while P
t
was set to be 0.12 kPa. In the experiment, the signal sampling
rate from the pressure sensors was 1000 Hz and a Median
filter with window size of 100 was applied to the signal. We
performed 15 trials for each soft object by placing them in
the center of the grippers for testing the repeatability of the
sensor.
HE et al.: SOFT FINGERTIPS WITH TACTILE SENSING AND ACTIVE DEFORMATION 5
Fig. 7. The figure shows the relative internal pressure change when feedback control is enabled with tactile sensors on the fingertips during the grasping test.
The results with 5 soft cylinders 1-5 with the diameter D0=10,20,30,40,50 mm are shown in sub-figures a-e, respectively, a standard deviation of 15
trials is shown. The y-axis label is same for sub-figures a-e.
Fig. 8. The relative internal pressure change of the soft objects during stable grasping. The result of tactile sensors on fingertips are shown with comparison
to results of set gripper positions D=0.9D0,0.8D0,0.7D0,0.6D0. The result with 5 soft cylinders 1-5 with the diameter D0=10,20,30,40,50 mm are
shown in sub-figures a-e, respectively,a standard deviation of 15 trials is shown.The y-axis label is same for sub-figures a-e.
For only the position control experiment, the input angle
θof the gripper was calculated with given distance D using
the bilateration method [29]. Two vectors p4,5and p4,3are
connecting point P
4to P
5and point P
4to P
3, as shown in Fig. 3.
These two vectors can form a bilateration matrix, Z4,5,3, and
then p4,3can be computed as:
p4,3=Z4,5,3p4,5(3)
where
Z4,5,3=1
2s4,5s4,3+s4,5s5,344,5,3
44,5,3s4,3+s4,5s5,3,(4)
and
4,5,3=±1
4q(s4,3+s4,5+s5,3)22(s2
4,3+s2
4,5+s2
5,3),(5)
with 4,5,3being the oriented area of the triangle defined by
points P
4,P
5, and P
3(4453). The positive or negative of it
implies that p4,3can point in one of two directions. When P
3
is to the right of vector p4,5, the oriented area is negative;
otherwise it is positive. With the same equation, P
2can be
computed with the same theory by points P
5,P
3, and P
2(4532):
P
2=P
5+Z5,3,2p5,3.(6)
A detailed description of these formulae can be found in [29],
[30].
C. Results
Fig. 7 shows the crushing test results of the fingertip
with tactile sensors. During grasping, the internal pressure
of the grasped soft objects exhibit a gradual increase until
the increased contact pressure triggered the controller to
stabilize. For all the trials, the sensor shows good consistency
and repeatability with a average standard deviation error
of 0.24, 0.34, 0.42, 0.25, and 0.08 kPa for soft cylinders
D0=10,20,30,40,50 mm, respectively. When the fingertips
interact with a soft object during grasping tasks, the relative
sensitivity of the tactile sensors is affected by the stiffness
of the fingertips and the stiffness of the soft object. When the
stiffness of the fingertips is fixed, as shown in the experiments,
the tactile sensors are more sensitive when the object is
stiffer. In the study, although all of the soft cylinders are 3D
printed with the same material and same thickness, the smaller
diameter cylinders would have higher strain energy density
due to the structure. Thus, smaller cylinders exhibit higher
stiffness. On the other hand, smaller volume soft cylinders
would cause a larger pressure change with the same amount
of deformation (PV =nRT where n,R,Tare constants).
Considering the two factors (structure stiffness and volume),
soft cylinder 2 D0=20mm exhibits the highest pressure
variation during grasping in the experiment.
The stabilized pressure data from the soft objects with
closing gripper distance D=0.9D0,0.8D0,0.7D0,0.6D0
are compared with the test when tactile sensors are used for
control with pressure limit sets between P
s=0.1 kPa and
6 IEEE ROBOTICS AND AUTOMATION LETTERS. PREPRINT VERSION. ACCEPTED JANUARY, 2020
Fig. 9. (a) Experiment setup for the escaping test. The robot arm only moves
up and down with gradually increased speed and acceleration after the object
is secured in hand. (b) Two 0.5 N standard calibration weights were attached
on the soft object.
TABLE I
SET P RES SU RES F OR TH E ES CAP IN G TES T
Soft object diameter D0[mm] 10 20 30 40 50
Set pressure [kPa] 2 1.2 1.5 1 0.5
P
t=0.12 kPa. The relative internal pressure comparison results
between fingertips with and without tactile sensors are shown
in Fig. 8. It proves that deformation can be kept below 0.7
of the original diameter for all 5 different sized soft cylinders
under current set limit conditions. The set limit condition P
s
and P
tcan be tuned for lower or higher force grasp. The
sensitivity of the fingertips can also be increased by reducing
the material stiffness.
IV. ESCAPING TEST WITH HIGH-SP EE D MANIPULATION
A. Experiment setup
In the industrial environment, products are commonly
picked and placed from one location to another location.
Grasping the product with a gripper and ensuring the product
“follows” the trajectory of the manipulator is essential. With
only force closure by contact friction, the grasping normally
requires relatively high force, which makes it inappropriate
for soft and fragile products. By having form closure or
caging grasps, the capability of the object to “follow” the
manipulator without escaping can be significantly increased.
In this experiment, we attach the gripper on a UR5 robot arm
(Universal Robots, Denmark) to test the performance of the
fingertips during high-speed manipulations. We expect the type
2 fingertips with active deformation can secure the grasping
object better under high-speed manipulation compare to type
1 fingertips with only silicone pads.
Fig. 9 shows the setup of the experiment. The robot arm
was programmed to do a sinusoidal movement with gradually
increased speed and acceleration after the gripper grasped
the object and secured it in hand. Type 2 fingertips with
active shaping changing ability and type 1 fingertips with only
silicone pads were tested separately with soft objects with the
diameter D0=10,20,30,40,50 mm. 0.5 N standard calibration
a.
b.
c.
Fig. 10. (a) Gripper position during the escaping test. (b) Gripper acceleration
during the escaping test. (c) Relative moving distance between the gripper and
object during the escaping test. Red and blue cross marks show the position
and acceleration where the object loss in contact with type 1 fingertips and
type 2 fingertips, respectively. 15 trials for each fingertip are shown.
weights were attached on each side of the soft object to add ex-
tra initial inertia (shown in Fig. 9 b). Relative moving distance
calculated based on the distance change between the gripper
and object were used to evaluate the grasping robustness.
Motion tracking cameras (Flex 3 OptiTrack, NaturalPoint, Inc.,
USA) were used to record the gripper and object position
during manipulations with pre-attached reference markers. The
internal pressure change of the soft object was recorded and
used as a reference to ensure equal force was applied to
the object between type 1 and type 2 fingertips. Before each
manipulation, type 2 fingertip cavities were inflated to a certain
size to adapt to the object shape until the soft object reached
the set pressure. Active shape-changing of the fingertips is
needed to ensure the form-closure created for objects with
different diameters. Table I shows the set pressures of the 5
soft objects in the test. Each soft object was tested 3 times for
repetition.
B. Result
Fig. 10 a. shows the gripper position represented by the
end-effector of the UR5 robot arm as they move together as
a rigid body. Fig. 10 b. shows the acceleration of the gripper
during the test. Fig. 10 c. shows the relative moving distance
between the object and the gripper with 15 trials of test for
HE et al.: SOFT FINGERTIPS WITH TACTILE SENSING AND ACTIVE DEFORMATION 7
Fig. 11. (a) Form-closure model of the inflated fingertips with 4 Kel vin Voight models. (b) Example sensing data from the 4 pressurized soft cavities L1,
L2, R1, and R2 for the time period of 40 s to 55 s of one trial of soft object D0=30 mm test. Moving trajectory and acceleration during the period are
shown in Fig. 10.
pure silicone fingertips shown in red and dual-soft fingertips
shown in blue for the 5 different soft objects. When the relative
distance between the object and gripper is greater than 5 mm,
the object will be considered as “escape” during the test. All
the tests carried out with fingertips type 2, with active changed
shape, successfully secured the object during the high-speed
manipulation test with accelerations more than 5 m/s2. Only
1 out of the 15 trials experiment for fingertip type 2 had
displacement larger than 5 mm. In contrast, none of the 15
trials with fingertips type 1 passed the test, while the object
started to show “offset” when the gripper starts moving and
“loss contact” at manipulation accelerations as low as 0.1 m/s2.
The positions where the “escape” happened are marked with
cross marks in Fig. 10 a. and b. A clear oscillation can be
observed from the relative moving distance data, as shown in
Fig. 10 c.
During the manipulation tasks with fingertips type 1, friction
is mainly used to secure the object in hand Ffmg =matand
Ff6µsN, where Ffis the friction between fingertips and the
object, mis the mass of the object, atis the object acceleration
in vertical direction, µsis the static coefficient of friction,
and Nis the applied normal force from the grasping. The
object starts losing contact with the gripper when its inertia
overcomes the static friction when the acceleration is high. The
object will be secured again when the gripper reduces its speed
when approaching the peak points. The oscillations observed
in Fig. 10 c. for type 1 fingertips manipulations reflects these
short-period of secured grasps.
For the manipulation tasks with fingertips type 2, small
oscillations can also be observed, particularly during high
acceleration periods. However, the oscillations are relatively
smaller in amplitudes compare to fingertips type 1, and the
causes for the oscillations of the two types of fingertips are
different. The form-closure structure formed by the active
shape-changing fingertips will also act as soft dampers to
reduce the contact force between the gripper and soft object.
Viscoelastic material such as silicone can be modelled by con-
necting a spring and a dashpot in parallel as Kelvin Voight
model [31]. Fig. 11a. shows the model of grasping with form-
closure fingertips. During high accelerations, the fingertips will
thus deform according to the direction and amplitude of the
acceleration. The form-closure structure will then behave like
a tight caging with shock absorbers which allow the object to
move during dynamic conditions. Extra protection of the soft
object can be provided with smaller contact force compared
to rigid form-closure grippers. This explains the oscillation of
the relative distance between the object and gripper for type
2 fingertips during high-acceleration manipulations.
In addition, the inflated fingertips can also sense the force
between the object and different area of the fingertips during
the manipulation. Fig. 11 b. shows the sensor reading from
L1, L2, R1, and R2 pressurized air cavities during a period
of manipulation. The data is shown by one trial of soft object
3 and D0=30 mm manipulation test for the period of 40
to 55 s. The sensor reading data reflects the oscillations
in the relative moving distance, as shown in Fig. 10 c.
Clear oscillations of the relative pressure can be observed
when higher and lower forces were applied to the object.
In the experiment, the cylindrical object was placed in the
center of the gripper. Due to the in-hand location of the
cylindrical object and the sinusoidal movement, the top two
pressurized air cavities L1 and R1 show similar trend while
the bottom two pressurized air cavities L2 and R2 show
opposite trends. When the object moves upward relative to
the gripper, the top two sensor readings increase while the
bottom two sensor readings decrease. The curved morphology
of the fingertips allows the tactile sensors to sense forces in
different configuration compared to its original non-inflated
states. The in-hand position and force distribution of the object
can, therefore, be monitored during the manipulation.
V. CONCLUSIONS
This paper presents the concept of a soft fingertip design
with the dual function of tactile sensing and active shape-
changing. The fingertips are integrated on a traditional grasper
with four-bar-linkage fingers based on the GR2 robot gripper,
with pressure feedback control to enable safe grasping with
soft and delicate objects. The proposed fingertips achieve
8 IEEE ROBOTICS AND AUTOMATION LETTERS. PREPRINT VERSION. ACCEPTED JANUARY, 2020
form closure by inflating different sections to change its
shape during grasping. With the soft form-closure structure,
the gripper can not only secure the object for high-speed
manipulation but also reduce the increased contact impact
force caused by inertia. In addition, when the fingertip is
inflated, the change of its morphology would change the way
the fingertip senses, from flat to a curved baseline.
In this study, we limited the objects to soft sensorised cylin-
drical objects with different diameters to clearly demonstrate
the dual function of active shape-changing and sensing of
the fingertip. Objects from the YCB grasping test were also
grasped with a success rate of 100% (repeated 3 times) to
show the grasping reliability and repeatability. Selected testing
videos are also included in the supplementary materials[32].
One limitation of the current design is that actuation of the
fingertip for shape-changing of a single air cavity cannot be
achieved simultaneously with its sensing function. Further
detailed calibrations of the fingertips and design of more
specific controllers are needed for future applications.
Rigid grippers have the advantages of low cost, simple
but precise control, and high reliability while soft robotic
grippers are able to safely interact with delicate objects due to
its high adaptability and low contact pressure. The fingertip
design we proposed combined the advantages of rigid and
soft robotic grippers by: 1) using rigid mechanism for precise
general grasping and manipulation, 2) applying soft gripper
actuation in a local contact level to retain safe interaction
with the object, and 3) integrating soft tactile sensing on
the fingertip for feedback control. Grippers with such dual-
function fingertips have the advantage to be used in food
industries with food products that are easy to bruise, tear, or
deform. Compared to the prehensile approach by most soft
robotic grippers, the form closure (which can behave like a
tight caging with shock absorbers) achieved by the fingertips
active shape-changing provides the potential to manipulate the
object without immobilizing it. Future study will focus on
using the shape-changing and tactile sensing ability of the
fingertip for dexterous in-hand manipulation.
REFERENCES
[1] P. Y. Chua, T. Ilschner, and D. G. Caldwell, “Robotic manipulation of
food products - a review,” Industrial Robot, vol. 30, pp. 345–354, 2003.
[2] F. Bader and S. Rahimifard, “Challenges for industrial robot applications
in food manufacturing,” ACM International Conference Proceeding
Series, 2018.
[3] S. Birrell, J. Hughes, J. Y. Cai, and F. Iida, “A field-tested robotic
harvesting system for iceberg lettuce,Journal of Field Robotics, vol. 0,
no. 0, 2019.
[4] J. Gray and S. Davis, “Robotics in the food industry: an introduction,” in
Robotics and Automation in the Food Industry, ser. Woodhead Publishing
Series in Food Science, Technology and Nutrition, D. G. Caldwell, Ed.
Woodhead Publishing, 2013, pp. 21 – 35.
[5] C. Blanes, M. Mellado, C. Ortiz, and A. Valera, “Review. technologies
for robot grippers in pick and place operations for fresh fruits and
vegetables,Spanish Journal of Agricultural Research, vol. 9, no. 4,
pp. 1130–1141, 2011.
[6] C. Blanes, M. Mellado, and P. Beltran, “Novel additive manufacturing
pneumatic actuators and mechanisms for food handling grippers,” Actu-
ators, vol. 3, no. 3, pp. 205–225, 2014.
[7] P. Pedro, C. Ananda, P. B. Rafael, A. R. Carlos, and B. C. Alexan-
dre, “Closed structure soft robotic gripper,2018 IEEE International
Conference on Soft Robotics, RoboSoft 2018, vol. 2, pp. 66–70, 2018.
[8] R. Sam and S. Nefti, “A new design approach of robotic gripper for
reducing operating cost for handling food product,” 2010 IEEE 9th
International Conference on Cybernetic Intelligent Systems, CIS 2010,
pp. 1–5, 2010.
[9] K. K. Chan and N. C. Cheung, “A novel two-finger variable-reluctance
gripper for high-speed grasping of delicate objects: An implementation
case study,” IEEE Transactions on Industrial Electronics, vol. 52, no. 6,
pp. 1705–1707, 2005.
[10] B. Ward-Cherrier, N. Rojas, and N. F. Lepora, “Model-Free Precise in-
Hand Manipulation with a 3D-Printed Tactile Gripper,IEEE Robotics
and Automation Letters, vol. 2, no. 4, pp. 2056–2063, 2017.
[11] F. Ilievski, A. D. Mazzeo, R. F. Shepherd, X. Chen, and G. M. White-
sides, “Soft robotics for chemists,” Angewandte Chemie - International
Edition, vol. 50, no. 8, pp. 1890–1895, 2011.
[12] E. Brown, N. Rodenberg, J. Amend, A. Mozeika, E. Steltz, M. R. Zakin,
H. Lipson, and H. M. Jaeger, “Universal Robotic Gripper based on the
Jamming of Granular Material,” 2010.
[13] Q. Lu and N. Rojas, “On soft fingertips for in-hand manipulation:
Modeling and implications for robot hand design,” IEEE Robotics and
Automation Letters, vol. 4, no. 3, pp. 2471–2478, 2019.
[14] Z. Wang, D. S. Chathuranga, and S. Hirai, “3D printed soft gripper for
automatic lunch box packing,” 2016 IEEE International Conference on
Robotics and Biomimetics, ROBIO 2016, pp. 503–508, 2016.
[15] A. Petterson, T. Ohlsson, D. G. Caldwell, S. Davis, J. O. Gray, and T. J.
Dodd, “A Bernoulli principle gripper for handling of planar and 3D
(food) products,” Industrial Robot, vol. 37, no. 6, pp. 518–526, 2010.
[16] V.-D. Nguyen, “Constructing force- closure grasps,” The International
Journal of Robotics Research, vol. 7, no. 3, pp. 3–16, 1988.
[17] A. Bicchi and C. E. Piaggio, “On the Closure Properties of Robotic
Grasping,” J. of Robotics Research, 1993.
[18] M. Honarpardaz, M. Tarkian, J. ¨
Olvander, and X. Feng, “Finger de-
sign automation for industrial robot grippers: A review,” Robotics and
Autonomous Systems, vol. 87, pp. 104–119, 2017.
[19] H. Song, M. Y. Wang, and K. Hang, “Fingertip Surface Optimization for
Robust Grasping on Contact Primitives,IEEE Robotics and Automation
Letters, vol. 3, no. 2, pp. 742–749, 2018.
[20] N. Herzig, P. Maiolino, F. Iida, and T. Nanayakkara, “A Variable
Stiffness Robotic Probe for Soft Tissue Palpation,IEEE Robotics and
Automation Letters, vol. 3, no. 2, pp. 1168–1175, 2018.
[21] C. Jackson, H. Vermaak, and G. Jordaan, “Force sensing for dynamic
gripping, using a piezoelectric sensor,” IEEE International Symposium
on Industrial Electronics, pp. 1371–1376, 2007.
[22] A. Petrovskaya, O. Khatib, S. Thrun, and A. Y. Ng, “Bayesian estimation
for autonomous object manipulation based on tactile sensors,” Proceed-
ings - IEEE International Conference on Robotics and Automation, vol.
2006, pp. 707–714, 2006.
[23] H. Fischer and R. Wolf, “Tactile pressure sensor,” Nov. 16 1999, uS
Patent 5,983,725.
[24] Y. Tenzer, L. P. Jentoft, and R. D. Howe, “The feel of mems barometers:
Inexpensive and easily customized tactile array sensors,IEEE Robotics
& Automation Magazine, vol. 21, no. 3, pp. 89–95, 2014.
[25] N. Rojas, R. R. Ma, and A. M. Dollar, “The GR2 Gripper: An Un-
deractuated Hand for Open-Loop In-Hand Planar Manipulation,” IEEE
Transactions on Robotics, vol. 32, no. 3, pp. 763–770, 2016.
[26] J. Kim, A. Alspach, and K. Yamane, “3d printed soft skin for safe
human-robot interaction,” in 2015 IEEE/RSJ International Conference
on Intelligent Robots and Systems. IEEE, 2015, pp. 2419–2425.
[27] Stratasys, “Digital Materials ( DMs ) Data Sheet,” 2013. [Online].
Available: http://stratasys.com/materials/polyjet/digital-materials
[28] T. Nanayakkara, A. Jiang, M. del Roc´
ıo Armas Fern´
andez, H. Liu,
K. Althoefer, and J. Bimbo, “Stable grip control on soft objects with
time-varying stiffness,IEEE Transactions on Robotics, vol. 32, no. 3,
pp. 626–637, June 2016.
[29] N. Rojas, “Distance-based formulations for the position analysis of kine-
matic chains,” Ph.D. dissertation, Universitat Polit`
ecnica de Catalunya,
2012.
[30] N. Baron, A. Philippides, and N. Rojas, “A Novel Kinematically
Redundant Planar Parallel Robot Manipulator With Full Rotatability,”
Journal of Mechanisms and Robotics, vol. 11, no. 1, 11 2018.
[31] I. T. Jaswal and A. Mahmood, “Robotic gripping using soft finger tips,
ICIEECT 2017 - International Conference on Innovations in Electrical
Engineering and Computational Technologies 2017, Proceedings, pp.
1–5, 2017.
[32] B. Calli, A. Walsman, A. Singh, S. Srinivasa, P. Abbeel, and A. M.
Dollar, “Benchmarking in manipulation research: Using the yale-cmu-
berkeley object and model set,IEEE Robotics & Automation Magazine,
vol. 22, no. 3, p. 36–52, Sep 2015.
... a soft pneumatic chamber was directly 3D-printed on the finger to detect tactile forces. The paper presented in [10] proposed a novel soft fingertip with embedded air cavities that can be used as a sensing element and actively change the shape of the fingertip, thus achieving a robust grasp while manipulating delicate objects. Although these solutions have demonstrated the benefits of introducing tactile feedback to a soft gripper or finger, they only allow the user to retrieve lumped contact information only. ...
... Conversely, our design is monolithic and can therefore be fully 3D printed without requiring further integration of the parts, simplifying the fabrication of the hand. The solutions provided in [7], [10], [14], [15] have successfully integrated soft tactile sensors on robotic grippers or hands, however the sensing is limited to fingertips. Our design utilises distributed sensing to provide information over the contact locations and pressures applied on the hand. ...
... Pneumatic chambers are chosen as the tactile sensing methodology in the hand design due to the inherent sensor softness, and ease of integration with single-batch 3D printing [10], [14]. In this respect, the sensors are designed as airfilled soft membranes that are raised 4mm from the surface of the fingers and palm, with 1.5mm thickness and material stiffness of Shore A40 hardness. ...
Article
Humans rely on distributed tactile sensing in their hands to achieve robust and dexterous manipulation of delicate objects. Soft robotic hands have received increased attention in recent years due to their adaptability to unknown objects and safe interactions with the environment. However, the integration of distributed sensing in soft robotic hands is lacking. This is largely due to the complexity in the integration of soft sensing solutions with the hands. This letter proposes a novel soft robotic hand that incorporates an active palm and distributed pneumatic tactile sensing in both the fingers and the palm. Multi-material 3D printing allows the tactile sensors to be directly printed on the hand, whereas conventional tactile approaches require the sensors to be attached as part of multiple fabrication procedures. Active degrees of freedom are introduced in the palm to achieve increased dexterity. The proposed hand successfully performed 32 of the 33 Feix taxonomy grasps and all 11 Kapandji thumb opposition poses.
... However, those sensors are facing difficulties in complex contact modelling, limited sampling rate, constraint geometry, and size. Another critical factor limiting the application of soft tactile sensors is their adaptability and robustness when interacting with unstructured environments 10 . A tactile sensor can be very sensitive for a particular low-force range but easily saturate and damage when there is an unexpected high force 3 . ...
... Actuation and perception can be considered as an integration 10,11 . When an object interacts with the environment, the tactile measurements ultimately depend on the sensor configurations, especially in soft sensors. ...
... Examining the tissue with different stiffness of the probe can significantly reduce the uncertainty of the measured haptic information, thus, more effective in the detection. While previous development of active sensing predominately relies on the stiffness and action change of the agent where the tactile sensor is mounted on a probe, coupled actuation and perception can also be incorporated in the sensor design itself 10 . This approach is particularly useful, considering that many robots interact with the environment directly with tactile sensors. ...
Article
Full-text available
The stiffness of a soft robot with structural cavities can be regulated by controlling the pressure of a fluid to render predictable changes in mechanical properties. When the soft robot interacts with the environment, the mediating fluid can also be considered an inherent information pathway for sensing. This approach to using structural tuning to improve the efficacy of a sensing task with specific states has not yet been well studied. A tunable stiffness soft sensor also renders task-relevant contact dynamics in soft robotic manipulation tasks. This paper proposes a type of adaptive soft sensor that can be directly 3D printed and controlled using pneumatic pressure. The tunability of such a sensor helps to adjust the sensing characteristics to better capturing specific tactile features, demonstrated by detecting texture with different frequencies. We present the design, modelling, Finite Element Simulation, and experimental characterisation of a single unit of such a tunable stiffness sensor. How the sensing characteristics are affected by adjusting its stiffness is studied in depth. In additional to the tunability, the results show such type of adaptive sensors exhibit good sensitivity (up to 2.6 [KPa/N]), high sensor repeatability (average std < 0.008 [KPa/N]), low hysteresis (< 6%), and good manufacturing repeatability (average std = 0.0662[KPa/N]).
... Tactile sensing at the external interface of robotic end-effectors enables a direct feedback loop to modulate applied force during interactions (Zou et al., 2017). The adaptability of soft tactile sensors enables the contact surface to conform to the shape of objects and distribute any handling forces over a larger contact area (He et al., 2020a). ...
... A wide range of tactile sensing technologies has been developed with successful designs coming from both close biomimicry as well as deviating from nature to incorporate additional capabilities (Zou et al., 2017;Chi et al., 2018). One such deviation proposed in literature comprises adjusting the effective stiffness of the soft tactile membrane via pneumatic actuation in order to allow for the sensitivity and measurement range of the sensor to be adjusted, leading to a more generically applicable sensor (He et al., 2020a;Jenkinson et al., 2020;Zhang et al., 2021). Previous studies exploring pneumatic actuation within soft tactile sensors have applied this to reactive grasping (McInroe et al., 2018), shape identification (Huang et al., 2019;Xiang et al., 2019), estimating tissue elastic modulus (Gubenko et al., 2017) and an explorative capsule capable of self-locomotion (Hinitt et al., 2015). ...
... Optics-based tactile sensing is common across soft actuated sensors, as all electrical components can be physically separated from the tactile membrane (Shimonomura, 2019). Among the pneumatically actuated tactile sensors referenced above, only He et al. (2020a) and He et al. (2020b) use pneumatic variations as the primary form of sensing; all other studies chose to implement optics-based tactile sensing. These studies have focused on detecting surface or bulk features of the stimulus that they are sensing, where heterogeneity in the depth of the stimulus remains under-exploited as a tactile cue. ...
Article
Full-text available
Soft tactile sensors are an attractive solution when robotic systems must interact with delicate objects in unstructured and obscured environments, such as most medical robotics applications. The soft nature of such a system increases both comfort and safety, while the addition of simultaneous soft active actuation provides additional features and can also improve the sensing range. This paper presents the development of a compact soft tactile sensor which is able to measure the profile of objects and, through an integrated pneumatic system, actuate and change the effective stiffness of its tactile contact surface. We report experimental results which demonstrate the sensor’s ability to detect lumps on the surface of objects or embedded within a silicone matrix. These results show the potential of this approach as a versatile method of tactile sensing with potential application in medical diagnosis.
... This is particularly challenging in the design of soft robots for extreme environment operations (e.g., nuclear plant exploration with high-radiation, tissue interaction during metal-free scenarios in magnetic resonance imaging). Although rigid electronic controllers allow multimodal sensor integration into the system [6], the conversion of sensory information into an electrical signal before being processed to pneumatic signal control brings additional system complexity and limits the application in extreme conditions. A potential solution is to replace the rigid pneumatic power transmission with soft computational elements, where the adaptability and usability of the soft robotic system can be improved. ...
Article
Although research studies in pneumatic soft robots develop rapidly, most pneumatic actuators are still controlled by rigid valves and conventional electronics. The existence of these rigid, electronic components sacrifices the compliance and adaptability of soft robots. Current electronics-free valve designs based on soft materials are facing challenges in behaviour consistency, design flexibility, and fabrication complexity. Taking advantages of soft material 3D printing, this paper presents a new design of a bi-stable pneumatic valve, which utilises two soft, pneumatically-driven, and symmetrically-oriented conical shells with structural bistability to stabilise and regulate the airflow. The critical pressure required to operate the valve can be adjusted by changing the design features of the soft bi-stable structure. Multi-material printing simplifies the valve fabrication, enhances the flexibility in design feature optimisations, and improves the system repeatability. In this work, both a theoretical model and physical experiments are introduced to examine the relationships between the critical operating pressure and the key design features. Results with valve characteristic tuning via material stiffness changing show better effectiveness compared to the change of geometry design features (demonstrated largest tunable critical pressure range from 15.3 to 65.2 kPa and fastest response time ≤ 1.8s).
... This is particularly *This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) Grant EP/V000748/1; S. Wang was supported by the CSC-PAG Oxford Scholarship for his DPhil study. 1 challenging in the design of soft robots for extreme environment operations (e.g., nuclear plant exploration with high-radiation, tissue interaction during metal-free scenarios in magnetic resonance imaging). Although rigid electronic controllers allow multimodal sensor integration into the system [6], the conversion of sensory information into an electrical signal before being processed to pneumatic signal control brings additional system complexity and limits the application in extreme conditions. A potential solution is to replace the rigid pneumatic power transmission with soft computational elements, where the adaptability and usability of the soft robotic system can be improved. ...
Preprint
Full-text available
Although research studies in pneumatic soft robots develop rapidly, most pneumatic actuators are still controlled by rigid valves and conventional electronics. The existence of these rigid, electronic components sacrifices the compliance and adaptability of soft robots.} Current electronics-free valve designs based on soft materials are facing challenges in behaviour consistency, design flexibility, and fabrication complexity. Taking advantages of soft material 3D printing, this paper presents a new design of a bi-stable pneumatic valve, which utilises two soft, pneumatically-driven, and symmetrically-oriented conical shells with structural bistability to stabilise and regulate the airflow. The critical pressure required to operate the valve can be adjusted by changing the design features of the soft bi-stable structure. Multi-material printing simplifies the valve fabrication, enhances the flexibility in design feature optimisations, and improves the system repeatability. In this work, both a theoretical model and physical experiments are introduced to examine the relationships between the critical operating pressure and the key design features. Results with valve characteristic tuning via material stiffness changing show better effectiveness compared to the change of geometry design features (demonstrated largest tunable critical pressure range from 15.3 to 65.2 kPa and fastest response time $\leq$ 1.8 s.
... Due to the compliance of elastomeric materials, novel soft robots can adapt their morphology to uncertain environments with the emerging cross-disciplinary research of material science, control, and rapid prototyping [1]. Soft robotic approach, therefore, show as a good supplement to the conventional rigid robot for application in Minimally Invasive Surgery [2], human interaction [3], soft and delicate object manipulation [4,5], and soft sensing [6,7]. ...
Article
Soft continuum robots require differential control of channel pressure across several modules to trace 3D trajectories at the tip. For current designs of such actuators, sheathing is required to prevent radial expansion when the chambers are pressurized. With the recent development of soft materials additive manufacturing, 3D printing has become a promising fabrication method for soft continuum robots. However, most current designs for continuum actuators are based on molding, which are not designed for 3D printing. This paper proposes an internal constraint-based soft continuum actuator for single material 3D printing, with tunable design parameters to render pre-defined motions. The internal constraint method maximizes the superiority of the rapid prototyping solution in terms of customizing the soft continuum actuators with high fabrication speed and design freedom. The internal constraints come in the form of internal beam elements that not only limit the undesired radial expansion (up to ∼14% of conventional design) but also allows the actuator to be pressurized at a higher driving pressure (up to ∼160%) and higher maximum bending angle (up to ∼320%) compared to conventional no-beam design. By tuning the design parameter q (determined by the number of constraint beams n per radial cross-section and the number of such sections k along the axial direction), we can render the actuator to the desired movement under specific driving pressure p. We show numerical simulation and hardware experiment results for a soft actuator to achieve specified bending and twisting motions following this design approach.
... Pneumatic inflation, on the other hand, is compact in control and flexible to be integrated into diverse systems [18]. However, the elastomeric material is prone to exhibit non-linear deformation when stiffness is controlled [19]. Conventional sensing methods focused on adding additional sensors in the actuation system. ...
Article
Robotic phantoms enable advanced physical examination training before using human patients. In this article, we present an abdominal phantom for palpation training with controllable stiffness liver nodules that can also sense palpation forces. The coupled sensing and actuation approach is achieved by pneumatic control of positive-granular jammed nodules for tunable stiffness. Soft sensing is done using the variation of internal pressure of the nodules under external forces. This article makes original contributions to extend the linear region of the neo-Hookean characteristic of the mechanical behavior of the nodules by 140% compared to no-jamming conditions and to propose a method using the organ level controllable nodules as sensors to estimate palpation position and force with a root-mean-square error of 4% and 6.5%, respectively. Compared to conventional soft sensors, the method allows the phantom to sense with no interference to the simulated physiological conditions when providing quantified feedback to trainees, and to enable training following current bare-hand examination protocols without the need to wear data gloves to collect data.
Article
We present a highly conformable stretchable sensory skin made of soft components. The skin uses low pressure pneumatic taxels and stretchable channels to conduct pressure signals to off-board micro-electromechanical systems (MEMS) pressure sensors. The force sensing range captures down to 0.01 N and has a roll-off of frequency of 102 Hz. Due to pressure equilibration, the signals are substantially unaffected as the channels stretch and flex over robot joints with up to 100% elongation. The design permits additional floating taxels which, though less insensitive to stretch, provide sensing of contacts across joints and gaps. With no conductive elements or rigid components, the skin is robust to impacts and withstands many cycles of stretch and immersion in water. We apply the skin to a robotic wrist with two intersecting axes and show that it has sufficient sensitivity and bandwidth to detect the onset of sliding as the robot contacts objects. We demonstrate the skin in object acquisition tasks in a constrained environment for which extraneous contacts are unavoidable.
Article
Full-text available
Agriculture provides an unique opportunity for the development of robotic systems; robots must be developed which can operate in harsh conditions and in highly uncertain and unknown environments. One particular challenge is performing manipulation for autonomous robotic harvesting. This paper describes recent and current work to automate the harvesting of iceberg lettuce. Unlike many other produce, iceberg is challenging to harvest as the crop is easily damaged by handling and is very hard to detect visually. A platform called Vegebot has been developed to enable the iterative development and field testing of the solution, which comprises of a vision system, custom end effector and software. To address the harvesting challenges posed by iceberg lettuce a bespoke vision and learning system has been developed which uses two integrated convolutional neural networks to achieve classification and localization. A custom end effector has been developed to allow damage free harvesting. To allow this end effector to achieve repeatable and consistent harvesting, a control method using force feedback allows detection of the ground. The system has been tested in the field, with experimental evidence gained which demonstrates the success of the vision system to localize and classify the lettuce, and the full integrated system to harvest lettuce. This study demonstrates how existing state‐of‐the art vision approaches can be applied to agricultural robotics, and mechanical systems can be developed which leverage the environmental constraints imposed in such environments.
Article
Full-text available
During abdominal palpation diagnosis, a medical practitioner would change the stiffness of their fingers in order to improve the detection of hard nodules or abnormalities in soft tissue to maximise haptic information gain via tendons. Our recent experiments using a controllable stiffness robotic probe representing a human finger also confirmed that such stiffness control in the finger can enhance the accuracy of detecting hard nodules in soft tissue. However, the limited range of stiffness achieved by the antagonistic springs variable stiffness joint subject to size constraints made it unsuitable for a wide range of physical examination scenarios spanning from breast to abdominal examination. In this paper, we present a new robotic probe based on a variable lever mechanism (VLM) able to achieve stiffness ranging from 0.64 N.m/rad to 1.06 N.m/rad, that extends the maximum stiffness by around 16 times and the stiffness range by 33 times. This paper presents the mechanical model of the novel probe, the Finite Element (FE) simulation as well as experimental characterization of the stiffness response for lever actuation.
Article
Full-text available
We address the problem of fingertip design by leveraging on the fact that most grasp contacts share a few classes of local geometries. In order to maximize the contact areas for achieving more robust grasps, we first define the concept of Contact Primitive, which represents a set of contacts of similar local geometries. Thereafter, we propose a uniform cost algorithm, which is formulated as a decision making process in a tree structure, to cluster a set of example grasp contacts into a finite set of Contact Primitives. We design fingertips by optimization to match the local geometry of each contact primitive, and then 3D print them using soft materials to compensate for optimization residuals. For novel objects, we provide an efficient algorithm to generate grasp contacts that match the fingertip geometries while together forming stable grasps. Comparing to a baseline of flat fingertip design, the experiment results show that our design significantly improves grasp stability and it is more robust against various uncertainties.
Article
Full-text available
The use of tactile feedback for precision manipulation in robotics still lags far behind human capabilities. This study has two principal aims: 1) To demonstrate in-hand reorientation of grasped objects through active tactile manipulation. 2) To present the development of a novel TacTip sensor and GR2 gripper platform for tactile manipulation. Through the use of Bayesian active perception algorithms, the system successfully achieved in-hand reorientation of cylinders of different diameters (20, 25, 30 and 35 mm) using tactile feedback. Average orientation errors along manipulation trajectories were below 5°. for all cylinders with reorientation ranges varying from 42°. to 67°. We also demonstrated an improvement in active tactile manipulation accuracy when using additional training data. Our methods for active tactile manipulation with the GR2 TacTip gripper are model-free, can be used to investigate principles of dexterous manipulation and could lead to essential advances in the areas of robotic tactile manipulation and tele-operated robots.
Article
Contact models for soft fingertips are able to precisely compute deformation when information about contact forces and object position is known, thus improving the traditional soft finger contact model. However, the functionality of these approaches for the study of in-hand manipulation with robot hands has been shown to be limited, since the location of the manipulated object is uncertain due to compliance and closed-loop constraints. This paper presents a novel, tractable approach for contact modelling of soft fingertips in within-hand dexterous manipulation settings. The proposed method is based on a relaxation of the kinematic equivalent of point contact with friction, modelling the interaction between fingertips and objects as joints with clearances rather than ideal instances, and then approximating clearances via affine arithmetic to facilitate computation. These ideas are introduced using planar manipulation to aid discussion, and are used to predict the reachable workspace of a two-fingered robot hand with fingertips of different hardness and geometry. Numerical and empirical experiments are conducted to analyse the effects of soft fingertips on manipulation operability; results demonstrate the functionality of the proposed approach, as well as a trade-off between hardness and depth in soft fingertips to achieve better manipulation performance of dexterous robot hands.
Conference Paper
The global food industry is facing many challenges due to the impact of climate change, ever-changing demands by consumers, and increasing legislative pressures by the government, which have resulted in several drivers for changes. Current large-scale rigid manufacturing systems are increasingly seen as incapable of supporting the underlining requirements for implementation of such changes. In this context, one of the key requirements is the need for improved flexibility and reconfigurability of production facilities, often provided by adoption of Industrial Robots in other manufacturing sectors. However, despite their recent technological advancements, in particular the advent of the 4th industrial revolution (Industry 4.0), and significant reduction in overall implementation cost over the last two decades, the uptake of industrial robots in food processing has been slow. This paper explores the application of industrial robots in food manufacturing, the benefits of their use and the challenges currently hindering their uptake.
Article
This paper presents a novel kinematically redundant planar parallel robot manipulator, which has full rotatability. The proposed robot manipulator has an architecture that corresponds to a fundamental truss, meaning that it does not contain internal rigid structures when the actuators are locked. This also implies that its rigidity is not inherited from more general architectures or resulting from the combination of other fundamental structures. The introduced topology is a departure from the standard 3-RPR (or 3-RRR) mechanism on which most kinematically redundant planar parallel robot manipulators are based. The robot manipulator consists of a moving platform that is connected to the base via two RRR legs and connected to a ternary link, which is joined to the base by a passive revolute joint, via two other RRR legs. The resulting robot mechanism is kinematically redundant, being able to avoid the production of singularities and having unlimited rotational capability. The inverse and forward kinematics analyses of this novel robot manipulator are derived using distance-based techniques, and the singularity analysis is performed using a geometric method based on the properties of instantaneous centers of rotation. An example robot mechanism is analyzed numerically and physically tested; and a test trajectory where the end effector completes a full cycle rotation is reported. A link to an online video recording of such a capability, along with the avoidance of singularities and a potential application, is also provided.
Conference Paper
This paper focuses on the grippers or end effectors that are attached with robots to interact with the environment. The analysis of the fundamental deformation of soft fingertips are discussed. In order to analyze the deformation effect, the dynamic relation of a five-fingered object manipulation using soft fingertips are explained. One way to grip the object with soft finger tips in a way that a viscoelastic material is attached at the tip of gripper. This viscoelastic material is then modelled as spring and damper and by including the friction between the finger tips and the object. We model this system using bond graph technique and then simulated in 20sim. Further we study the effects in variation of friction, damping and stiffness in simulation results, which shows that the deformation effect of soft fingertips during the gripping process is crucial for stable manipulation.