Design and development of the soft robotic gripper used

Journal of Mechanical Engineering Research and

Developments ISSN: 1024-1752

CODEN: JERDFO

Vol. 44, No. 3, pp. 334-345

Published Year 2021

334

Design and development of the soft robotic gripper used for the

food packaging system

Hoang Minh Dang†, Chi Thanh Vo†, Nhan Tran Trong†, Viet Duc Nguyen‡, Van Binh

Phung‡†*

† Industrial University of Ho Chi Minh City, Ho Chi Minh City, Vietnam ‡ Thuyloi University, 175 Tay Son, Dong Da, Hanoi, Vietnam

‡† Le Quy Don Technical University, Hanoi, Vietnam

*Corresponding author’s email: [email protected]

ABSTRACT

This paper presents the design and development of the soft robotic gripper with four actuators, which are made of

hyperelastic material and operated by pneumatic systems. Simulation of the gripper deformation and dynamics

issue was performed by using the finite element method in the ABAQUS software. The actuator was made by

casting method and the mold was 3D printed. The auxiliary mechanical device, which is integrated to the gripper,

is a two degree of freedom mechanism. It can move up and down and turn left and right conveniently. The

developed gripper can handle the object of different shapes with a geometrical limit up to 8 cm and gripping mass

up to 300 gram. Experimental results showed that the gripper can operate with the productivity of 3 objects per

minute that meets the actual requirements of the product packaging line in practice.

KEYWORDS

Soft robotic gripper, pneumatic actuator, hyperelastic material, design and development, food packaging system

INTRODUCTION

Soft robotics has been the potential area that attracts the attention of many researchers [1-3]. Movements of soft

robots are modeled in accordance with the organisms. One of the typical research objects in soft robotics is a soft

gripper that can handle items of any shape, indeterminate size, especially suitable when working with fragile

objects, e.g. glass, or easily-deformed objects, e.g. food. These are the outstanding advantages of the soft actuator

in comparison with the traditional robotic actuator. It can be implemented for many industries including food,

biology, medicine, ocean exploration, etc. [4].

The soft robotic gripper appeared first in the 1970s with the aim of emulating the functions of animal tendons [1],

then this research direction continued to expand and develop in the 1980s and in the 1990s new designs, new

control methods, as well as new materials, have been introduced [2, 3]. By the 2000s, thank to the emergence of

electrodynamic polymers, various kinds of soft robotic gripper were developed rapidly and a number of products

were already commercialized. Carozza et al. studied artificial hands using soft materials for the disabled [4]. A

typical study in this period was the gripper using a flexible mechanism, integrated with load cells, controlled via

cable developed by Dollar et al. [5]. In the meantime, Nakai et al. at the University of Tokyo studied a soft robot

prototype that can be configured, the robot structure is made of low hardness aluminium, creating the ability to

configuration transform thanks to the deformation of the structure [6].

So far, the soft robot gripper has been evolved in the direction of using hyperelastic material e.g. silicon [7-8].

The most common one is a soft gripper made of hyperelastic material, with a structure of hollow cavities

connecting to the pneumatic system [9-12]. Although the calculation and simulation of this type of gripper have

had a certain success, there are still some limitations when applied in practice, especially in terms of the relatively

small mass of the items that need to be gripped. In addition, this type of gripper is mostly a prototype under

experimental research form, the application in real production systems is still limited.

Design and development of the soft robotic gripper used for the food packaging system

335

3D Fingers Design:

- Shape, Size

- Wall Thickness

- Number of Compartments

- Number of fingers

Force simulation

Perform Regression Fn=f(h,P)

Designing Gripper Equipment

- Configuration of Lifting -

rotating - lowering mechanism

Manufacturing

Control System Design

Pick-up Experiment

ENDSize requirements

Feasible in

Manufacturing

and Control?

Satisfying

productivity?

Satisfy Force

Condition?

YES

NO

YES

NO

Pick-up Simulation

YES

NO

Figure 1. Procedure of design and development of the soft robotic gripper for the food packaging industry

The purpose of this paper is to introduce the design process for developing a soft gripper (consists of several

actuators or fingers), as illustrated in Figure 1, which is directly applied to the food packaging system. The

technical requirements are the maximum weight of the gripping object 300 gram, the gripping capacity of 3 objects

per minute. The shape of the object might flexibly be rectangular, spherical, and conical with a maximum profile

size of 8 cm. The gripping process requires that food or objects, in general, will not be damaged or crushed. These

requirements are built up from the actual need at food packaging factory with eggs, fruit, soft cakes, etc. and they

need to be placed into trays or boxes for packing and delivered to a large number of people such as places like

universities, schools, factories, etc.

Based on the requirement on the size of the gripping object, the 3D model of the gripper with four actuators is

built. Then, simulation is conducted to evaluate the gripping force of a finger on the object at different pressure

levels. Using the simulation data, the correlation between the gripping force upon the pneumatic pressure and the

distance from the finger to the object is developed. Next, the dynamic simulation of the gripping process is

performed. The simulation allows for evaluating the ability to grasp and hold objects of the gripper in order to

determine the design option that meets the requirements. Once a suitable design option is defined, the subsequent

steps including structural design, fabrication and validation of the prototype are carried out.

DESIGN OF THE SOFT ROBOTIC GRIPPER

Simulation of the actuator deformation


336

(a)(b)

(c) (d) (e)

(f) (g) (h)

(c)

Figure 2. 3D model of the gripper of four actuators: (a) and (b)-actuator, (c)-gripper

Each finger of the actuator is designed with the empty spaces connected together so that air can be pumped in, as

can be observed in Figure 2 (a) and (b). The size and number of compartments will affect the applied force as well

as the impact ability of the finger on the gripping object. How this influence can be studied based on the simulation

and experimental validation after the finger is fabricated. The 3D design model of the actuator is illustrated in

Figure 2(c). Figure 3 shows an illustrative result of air pressures effect on the strain state of a finger. The finger

simulation was performed by using hyperelastic material with many different hardness levels (from 20A, 25A,

30A, 35A, 40A, 45A) in order to define the suitable hardness material.

Fix

Applied pressure

(a) (b)

Figure 3. Effect of air pressure on the strain state of an actuator: (a) boundary condition; (b) strain state

Determination of force interaction between gripper and object

The most important parameter defining the working ability of the gripper is the force exerted on the object. The

exerting force is characterized by 2 normal components (Fn and Ft), in which the Fn can be defined by using the

simulation and the Ft is determined in accordance with Fn and the coefficient of friction μ between the finger and

the object. Coefficient μ is obtained from the experimental methods. The simulation procedure is described in

Figure 4.


337

(a) (b) (c)

(d)

(e)

Figure 4. Simulation procedure of exerting force on the object from the finger

The correlation between Fn upon the pressure P and the distance from the gripper to the finger h is studied by

using the ABAQUS software. The Yeoh material model is used for the calculation. The finger model is meshed

with a tetrahedral C3D10H element, while the object model uses the C3D8R. The interaction of adjacent thin

walls is also taken into account to ensure the simulation results are similar to reality. The simulation process starts

from a 3D model of the finger (Figure 4a), then sets the boundary and load conditions (Figure 4b), then meshes

the model (Figure 4c) and finally run the computing program (Figure 4d). The calculation result of the Fn in

accordance with the different values of pressure P and distance h are shown in Table 1. Besides, the outcome is

also represented visually in Figure 5. From the experimental data, it is possible to build an approximate function

of Fn by using the least squares regression algorithm as follows:

( ) ( ) ( )

( )

( )

2 2

00 10 01 20 11 02

3 2 2 3

30 21 12 03

4 3 2 2 3 4

40 31 22 13 04

,nF h P p p h p P p h p hP p P

p h p h P p hP p P

p h p h P p h P p hP p P

= + + + + + +

+ + + + +

+ + + + +

(1)

Where, p00=0.03888; p10=0.0007751; p01=–0.002712; p20=–0.0003192; p11=0.0001123; p02=0.0004033;

p30=1.436e-005; p21=–1.039e-005; p12=–1.552e-006; p03=–6.734e-006; p40=–1.775e-007; p31=1.353e-007;

p22=8.847e-009; p13=1.051e-008; p04= 4.234e-008.

Table 1. The obtained result of Fn in accordance with the distance h and pressure P

h (mm)

P (kPa)


338

Figure 5. Range of values to correlate Fn, h and P

The obtained function Fn(h,P) makes a firm basis for estimating the object mass corresponding to the different

pressure P and distance h. Also, the function Fn allows for defining the required pressure P for gripping the object.

Dynamic simulation of grasp process

To evaluate the working ability of the gripper with the objects of different shapes, the dynamic simulation of the

grasping process is carried out taking into consideration the objects with three typical shapes such as rectangular,

sphere and cone, as presented in Table 2. These are also common shapes for grasping objects in the packaging

system [13]. The simulation of the gripping process is performed by using the Abaqus/Implicit module. The four

main steps to be taken are 3D modelling, boundary and load application, mesh, and analysis. The coefficient of

friction μ = 0.3 is considered. The result shows that the air pressure P=55 kPa, and the fingers, which are made of

hyperelastic material with hardness 35A, are capable of gripping the object of 300 gram as the maximum mass.

Table 2. Simulation steps

Step Rectangular object Spherical object Conical object

1. 3D

modelling

2. Load

application,

boundary

condition


339

3. Mesh

4. Result

analysis

Fabrication of the actuator

There are two principal methods of fabricating soft grippers such as 3D printing by using hyperelastic material

and casting. 3D printing is available for fabricating the gripper of complex shape, but it is quite costly. In this

study, the gripper is made by the casting method, the process of which is shown in Figure 6. Top, bottom and lid

molds are made by using the 3D printer, as shown in Figure 6 (a), (b), and (c) respectively. The process of pouring

the material, hardening and finishing process of the gripper fabrication are shown in Figure (d), (e), and (f)

respectively. Eventually, the actuators are assembled to make the entire gripper, as illustrated in Figure 7. (a)

(b)

(a) (b) (c)

(d) (e) (f)

Figure 6. Fabrication process of an actuator

Figure 7. Assembly of the gripper


340

VALIDATION OF THE SOFT ROBOTIC GRIPPER

Fabrication of the auxiliary mechanism and control system

In order to validate the soft robotic gripper to be used for the food packaging system, an auxiliary mechanical

device that integrated with the gripper was prepared. It is a two degree of freedom mechanism, as illustrated in

Figure 8. The operating principle of the device is as follows: When it needs to pick the object or perform the

gripping process, the air is pumped into the pneumatic tank and then pumped into the gripper with four grippers.

Air pressure is measured by using a sensor. Once the intake air pressure reaches the pre-set magnitude, the

solenoid valve will automatically disconnect and the compressed air pressure inside the gripper is kept constant

during the gripping process. When it needs to drop the object, the solenoid valve opens and the air will be released

from the gripper out. To ensure sufficient air supply in the overall operation cycle, the tank is equipped with a

self-supply unit. When the air in the tank decreases, the device controls to re-inflate the tank, reaching a fixed

magnitude of 5 kgf/cm2.

A cluster

of soft

grippers

Pressure

Gauges

Translational

Mechanism

Rotating

mechanism

Stepper

motor

Figure 8. Two degree of freedom mechanism

The control system is also established to manage the device. Diagrama of the motor and air-pump control system

is shown in Figure 9 and Figure 10 respectively.

+5V

+5V

+5V

+5V

+5V+12V

+12V

+12V

DRIVER TB6600

TRANSLATIONAL

MOTION CONTROL

MOTOR

ROTATIONAL MOTION

CONTROL MOTOR

ARDUINO R3

PRESSURE REDUCING MODULE

12V-5V DC

MODULE

BLUETOOTH

ARDUINO NANO

POWER

SWITCH

EMERGENCY

SWITCH

STARTDOWN LEFT RIGHTUP STOP

+12V

GAS

CONTROL

CIRCUIT

POWER SUPPLY

12V DC

Figure 9. Diagrama of the motor control system


341

SINGLE-DIRECTIONAL

ELECTRIC VALVE

PRESSURE

REGULATOR BUTTON

PRESSURE SENSORSARDUINO R3

EXHAUST GAS

AIR PUMP

+12V

+5V

A CLUSTER OF SOFT

GRIPPERS

MOTOR

CONTROL

CIRCUIT

Figure 10. Diagrama of the air-pump control system

Validation and analysis

To evaluate the working capability of the gripper and auxiliary mechanical device, a series of experimental pick-

up are performed. Different types of objects including food and fruits with mass varying from 28 gram to 300

gram. The experimental tests on the gripping process of different objects are presented in Table 3. The

experimental result shows that with the required air pressure of 55 ÷ 70 kPa, the developed soft robotic gripper

can handle objects with a mass of up to 300 gram. The object that needs to be picked can be in different types of

configuration. Also, the gripping process does not affect the food quality, for instance, it does not spoil the fruit.

In particular, the productivity of the grasping process is up to 3 objects per minute, which is suitable for application

in the food packaging system [13].

Table 3. Experimental tests on the gripping process of different objects

Object Mass Experimental Test

Croissant


342

Passion fruit

Apple

Yogurt

Onion


343

Baby Pumpkin

Tomato

Potato

Radicchio


344

Guava

CONCLUSION

The design and development of the soft robotic gripper with four actuators was carried out in this study. The

actuator was made of hyperelastic material and operated by using pneumatic systems. The developed gripper can

handle the object of different shapes with a geometrical limit up to 8 cm and gripping mass up to 300 gram. The

gripper was integrated with the auxiliary mechanical device to be used for the food packaging system.

Experimental results showed that the gripper can operate with the productivity of 3 objects per minute that meet

the actual requirements of the product packaging line in practice. The successful development of the gripper from

this study promises a bright future of upcoming more useful devices.

ACKNOWLEDGEMENT

The contribution to this paper and financial support from Industrial University of Ho Chi Minh City, Vietnam

under the research project № 20/1.1CK05 toghether with the contract № 20/HD-DHCN are highly acknowledged.

REFERENCES

[1] S. Hirose and Y. Umetani, “The Development of a Soft Gripper for the Versatile Robot Hand”, Mechanism

and Machine Theory, Vol. 13, Pp. 351-359, 1978.

[2] K. Suzumori, S. Iikura and H. Tanaka, “Flexible microactuator for miniature robots”, in the Proceedings

IEEE Micro Electro Mechanical Systems, 1991.

[3] B.C. Yoseph, T. Xue, M. Shahinpoor, J. Simpson and J. Smith, “New Mexico Flexible , Low-mass Robotic

Arm Actuated by Electroactive Polymers and Operated Equivalently to Human Arm and Hand”, In the

Proceedings of the Robotics 98: The 3rd Conference and Exposition/ Demonstration on Robotics for

Challenging Environments Sponsored by American Society of Civil Engineers, 1998.

[4] M.C. Carozza, G. Cappiello, G. Stellin, F. Zaccone, F. Vecchi, S. Micera and P. Dario, “On the development

of a novel adaptive prosthetic hand with compliant joints: experimental platform and EMG control”, In the

Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems, 2005.

[5] A.M. Dollar and R.D. Howe, “A robust compliant grasper via shape deposition manufacturing”. IEEE/ASME

Transactions on Mechatronics, Vol. 11, No. 2, Pp. 154-161, 2006.

[6] H. Nakai, Y. Kuniyoshi, M. Inaba and H. Inoue, “Metamorphic robot made of low melting point alloy”, In

the Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems, 2002.

[7] F. Ilievski, A.D. Mazzeo, R.F. Shepherd, X. Chen and G.M. Whitesides, “Soft Robotics for Chemists”,

Communication, Vol. 50, Pp. 1890–1895, 2011.

[8] 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”, In the Proceedings of the National

Academy of Sciences, Vol. 107, No. 44, Pp. 18809-18814, 2010.


345

[9] K.C. Galloway, K.P. Becker, B. Phillips, J. Kirby, S. Licht, D. Tchernov, R.J. Wood, and D.F. Gruber

Galloway, “Soft Robotic Grippers for Biological Sampling on Deep Reefs”, Soft robotics, Vol. 3, No. 1, Pp.

23-33, 2016.

[10] W. Hu, R. Mutlu, W. Li and G. Alici, “A Structural Optimisation Method for a Soft Pneumatic Actuator”,

Robotics, Vol. 7, No. 2, Pp. 24, 2018.

[11] B. Wang, A. McDaid, T. Giffney, M. Biglari-Abhari, and K.C. Aw, “Design, modelling and simulation of

soft grippers using new bimorph pneumatic bending actuators”, Cogent Engineering, Vol. 4, No. 1, 2017.

[12] F. Yanqiong, W. Jiangbei, and P. Wu, “A Novel Fabric-Based Versatile and Stiffness-Tunable Soft Gripper

Integrating Soft Pneumatic Fingers and Wrist”, Soft Robot, Vol. 6, No. 1, Pp. 1-20, 2019.

[13] Z. Wang and S. Hirai, “Chamber Dimension Optimization of A Bellow-Type Soft Actuator for Food

Material Handling”, in the Proceedings of IEEE International Conference on Soft Robotics (RoboSoft), Pp.

382-387, 2018.

Documents

Design and development of the soft robotic gripper used