Gripper Mechanisms: Fiber-Reinforced Soft Gripper

Gripper Mechanisms
Fiber-Reinforced Soft Gripper
Soft grippers offer numerous benefits over their more-rigid counterparts. Due to the inherent

cushioning of pneumatics they can conform to surfaces or objects, absorb energy to maintain
stability, exhibit physical robustness and human-safe operation at potentially low cost.
Fig. 1
Literature Review
Soft Robotics
Fiber-reinforced actuators are a class of soft actuator. The basic design consists of an
elastomer bladder wrapped with inextensible reinforcements. The fiber-reinforced bending
actuator documented here consists of a core bladder reinforced with a strain-limiting layer
and inextensible fibers. Material stiffness defines the amount of pressure required to actuate
and the output forces attainable. Low durometer (high strain) materials will bend at lower
pressures but will produce lower output forces.
Fig 2 shows the components of the actuator.
1.The actuator core is a hollow semi-cylindrical tube molded out of elastomer.

2. The inextensible, strain-limiting layer is a sheet of fiberglass attached to the flat face of the
actuator core, preventing that face from lengthening when inflated.
3. The entire actuator is wrapped with inextensible Kevlar thread, which restricts radial
expansion.
4. This thread wrapping is secured in place by molding an additional outer "skin" layer which
encapsulates the threads and the actuator core.
5. Finally, the actuator is plugged at both ends, forming a closed chamber. A vented screw is
installed in one of the end caps to provide access for an air source to inflate the chamber.
Variation in Motion
Bending Extending
Twisting Twisting & Extending
Fig. 2
Figure 3 shows the arrangement of air pump and valves for the actuation of soft robots in order
to carry out inflation and deflation cycles. As shown, it consists of an air pump to act as the
pressure source; two solenoid valves (one acts as an air supply valve while the other as an
exhaust valve) and a pressure sensor to measure the air pressure in the soft actuator.
Compressed air from the pump passes through the air supply solenoid valve and changes to
output pressure when the air supply solenoid valve turns ON. In this way, air from the supply
pump passes through the air supply solenoid valve and changes to output pressure. A PWM
(Pulse-Width Modulated) output is then produced on the output pin of the MCU to switch
ON/OFF the exhaust valve in order to produce an output pressure equal to the desired pressure.
The exhaust valve is also used to deflate the soft actuator - this is essential for crawling motions
of soft robots. The output pressure is fed back to the microcontroller via the pressure sensor. This
is to check if the desired pressure has become equal to the output pressure. Pressure corrections
then occur to produce an output pressure that is equal to the set pressure. Once the pressure
sensor has sensed that the desired pressure is equal to the output pressure, the exhaust valve will
turn OFF (close) in order to maintain a constant pressure
Fig. 3 Fig. 4
Print-in Place Gripper
A flexible filament like TPU is used to 3D print the gripper. A servo is used to actuate the
gripper by creating tension in the strings attached to the claw fingers. This is an economical
alternative to the soft robotic gripper but the gripping range is lesser than the former.
Fig. 5
Servo Gripper:
A servo-driven aluminum-made claw, which can be precisely controlled from the input PWM
signal. Provides more sturdiness and is more reliable in the longer run.
Fig. 6
Gripper Selection
Although soft gripper provides various advantages of adapting to the object to be lifted and low-
power but the high cost of flexible liquid silicone rubber and control board made us prefer the
servo gripper, which has a more durable claw and the only component with a probability of
failure is the servo which can be replaced easily. Print-in place gripper is more susceptible to
printing failure due to TPU and servo gripper provides more modularity as degrees of freedom
can be increased by attaching additional servos on a pan/tilt bracket.
Motor Selection
Model: MG995
Weight: 55 gm
Operating voltage: 4.8V~ 7.2V
Stall torque @4.8V : 13kg-cm
Stall torque @6.6V : 15kg-cm
The required object to lift, i.e., a golf weighing around 200

grams along with the frame on which the claw is attached
is to be lifted by the servos. Fig. 7 (a)
Weight of the frame along with claw: 400 grams

Total Load to be lifted: 400 + 200 = 600 grams.
The task requires the gripper to take hold of the ball
lying in a tub of 30 cm in depth.
Thus, the gripper is designed for a length of 30 cm.
Considering maximum load condition,
torque required to lift the gripper:
τ =r∗F
where, r is the effective length and F is the load.
Required Torque = (30) (0.6) = 18 kg-cm
Considering a safety of factor of 1.5,
required torque = (1.5) (18) = 27 kg-cm
Hence two servos of below specifications are chosen.
Figure 7(b) shows the frame on which the claw is attached. Fig. 7 (b)
Servo Gripper Selection
Paw material: Hard aluminum alloy

- Paw weight: about 120g (without servo)
- Maximum paw spacing: about 160mm
- Overall length of paw: approx. 112mm
(whole longest length when claws are closed)
- Overall width of paw: about 60mm
(the overall width when the paw is closed)
- Overall thickness of the paw: approx. 69mm (thickest
overall width of the paw)
Fig. 8
An eight-fingered claw is chosen to provide more grip and
resistance to forces due to flow while the AUV is in motion.
Mobile Application:
The gripper can also be controlled using voice commands detected
by integrating the google assistant with the application. A
Bluetooth module, HC-05, attached to Arduino Uno, which
controls the servos for movement is used to send and receive data
from mobile phone to the Arduino.
Fig. 9
Image Processing
Vision is essential for the path planning of the AUV and mapping the surrounding environment.
The vehicle is equipped with a Logitech C270 USB camera in front of the main hull providing
visual feed to plan its motion according to various tasks to be performed and which are lined up
in the mission planner of the AUV. Jetson Nano is the GPU which takes care of all the
processing tasks and also interfaces various sensors with Microprocessor unit, which is Arduino
Mega.
The usb_cam package is used extract the frames from the camera, extracting 24 frames per
second. The OpenCV library is used for computer vision applications using Python3 as a high-
level language.
The below steps are performed in order to extract the objects from the captured frame:
 Converting the RGB image into Grayscale

 Gaussian Blur: It is applied typically to reduce image noise and reduce detail. Since the
Fourier transform of a Gaussian is another Gaussian, applying a Gaussian blur has the
effect of reducing the image's high-frequency components. So blurring, we reduce the
edge content and makes the transition from one color to the other very smooth.
 Adaptive mean thresholding
 Morphological Opening: The opening operation erodes an image and then dilates the
eroded image, using the same structuring element for both operations. Morphological
opening is useful for removing small objects from an image while preserving the shape
and size of larger objects in the image.
 Canny Edge Detection: To detect various edges in the image.
 Contour Detection: All the closed contours in the frame are detected. For detection of
gate, the contour is detected by the maximum contour area and the aspect ratio condition.
The buoy is detected using the aspect ratio and the
 Maximum area extraction from the detected contours, which corresponds to the gate.
 The objects are detected using two different approaches:
 Method 1: A rectangle bounding the gate and buoy are extracted and drawn on the
frame using the Bounding Rectangle class based on aspect ratio.
 Method 2: A rectangle enclosing the gate and buoy are extracted and drawn on the
frame using the Rotated Rectangle class based on aspect ratio and the angle
subtended by it with respect to the horizontal axis.
The appropriate values of the area, aspect ratio and the angle are obtained from plots
depicting them and filtering out the optimum values for detection. Fig
 Centroid Detection: The centroid of the detected objects is found by calculating the
moments and the coordinates with respect to the frame are found.
 Calculating the centroid coordinate offsets from the center of the frame. The offset value
obtained in number of pixels is converted to metric units by using the focal length of the
camera as a factor. Thus, it sends the relevant information in the form a csv file, the
information from which can be extracted to send the proper thrust value for motion
control of the AUV.
Gate Detection
Fig. 10 (a) Original frame Fig. 10 (b) Gate detected from frame
Fig. 10 (c) Position offsets in vertical plane Fig. 10 (d) Area of the gate detected from the
(x and y corresponding to sway and heave frame. This value is used as a threshold for
respectively) from the center of the frame extraction.
captured.
Buoy Detection
Fig. 11(a) Original image frame showing

the buoy to be detected and avoided before
passing through the gate.
Fig.11(b) Buoy detected using rectangular

bounding box method.
Fig. 11(c) Buoy detected using the Rotated
rectangle method.
Fig. 12 (a) Fig. 12 (b)
Fig. 12 (c) Fig. 12 (d)

Fig. 12 (e) Fig. 12 (f)
From figure 12, figures (a) and (b) represent the position offsets and area detected of the buoy
from the frame using the bounding box method. Similarly, figures (c) and (d) represent the same
for rotated rectangle method for a different frame. Figures (e) and (f) represent the aspect ratio
and angle values which are used as a threshold for detection.

Gripper Mechanisms: Fiber-Reinforced Soft Gripper

Uploaded by

Copyright:

Available Formats

Gripper Mechanisms: Fiber-Reinforced Soft Gripper

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gripper Mechanisms: Fiber-Reinforced Soft Gripper

Uploaded by

Copyright:

Available Formats

Gripper Mechanisms

Fiber-Reinforced Soft Gripper

Soft grippers offer numerous benefits over their more-rigid counterparts. Due to the inherent

1.The actuator core is a hollow semi-cylindrical tube molded out of elastomer.

Twisting Twisting & Extending

Print-in Place Gripper

The required object to lift, i.e., a golf weighing around 200

Weight of the frame along with claw: 400 grams

Servo Gripper Selection

Paw material: Hard aluminum alloy

 Converting the RGB image into Grayscale

Fig. 11(a) Original image frame showing

Fig.11(b) Buoy detected using rectangular

Fig. 12 (a) Fig. 12 (b)

Fig. 12 (c) Fig. 12 (d)

You might also like