research article asymmetric bellow flexible pneumatic...

Research ArticleAsymmetric Bellow Flexible Pneumatic Actuator forMiniature Robotic Soft Gripper

Ganesha Udupa1 Pramod Sreedharan1 P Sai Dinesh2 and Doik Kim3

1Department of Mechanical Engineering Amrita Vishwa Vidyapeetham (Amrita University) Kollam Kerala 690525 India2Department of Electronics and Communication Amrita Vishwa Vidyapeetham (Amrita University) Kollam Kerala 690525 India3Interaction amp Robotics Research Center Korea Institute of Science and Technology Seoul 136-791 Republic of Korea

Correspondence should be addressed to Ganesha Udupa ammasganeshgmailcom

Received 23 May 2014 Revised 24 September 2014 Accepted 23 October 2014 Published 3 December 2014

Academic Editor Anthony A Maciejewski

Copyright copy 2014 Ganesha Udupa et alThis is an open access article distributed under the Creative CommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The necessity of the soft gripping devices is increasing day-by-day in medical robotics especially when safe gentle motions andsoft touch are necessary In this paper a novel asymmetric bellow flexible pneumatic actuator (AFPA) has been designed andfabricated to construct a miniaturised soft gripper that could be used to grip small objects The model of AFPA is designed usingsolid works and its bending motion is simulated in Abaqus software for optimisation and compared with experimental resultsThe actuator is fabricated using compression molding process that includes micromachining of the molds Experiments conductedshow the bending characteristics of the actuator at different pressures The actuator shows excellent bending performance and theeccentricity in its design supports increased bending or curling motion up to a certain extent compared to normal bellows withouteccentricity The effects of profile shape and eccentricity on the actuator performance are analysed and the results are presented

1 Introduction

Different types of soft actuators are developed that could gen-erate the bendingmotion by themselves or by themechanismthat bends due to their actuation [1 2] But most of theseactuators are not single chamber or the actuator as a singlecannot bend but a combination of them with appropriatemechanism does the necessary bending motion [3ndash5] Flex-ible pneumatic actuator (FPA) was designed by Joseph LMcKibben in the 1950s known as pneumatic muscle actuatorToshiba Corp (Japan) developed a three-degree-of-freedomactuator known as flexible microactuator (FMA) [6] Eventhough the FMAs with two or more chambers with fiberreinforcement provide multiple DOF they require multiplepressure supplies valves and sensors as well as complicatedmanufacturing Asymmetric flexible pneumatic actuators(AFPAs) have been developed for the first time during the1990s using asymmetric polymerrubber tube and rubberbellow actuators with proper reinforcement to overcome thedisadvantages of FMA and FPAs and proposed as an innova-tive method of fabricating a dexterous human hand [7 8] It

has also been applied to fabricate a four fingered robot gripper[7ndash9] and microwalking robot [10] The design and analysisfor application to robotic hand using asymmetric nitrilerubber actuators have been studied [11ndash13] Later symmetricthickness bellow actuators for miniature gripper fabricatedby moulding technique and rubber bonding process withexcimer light [14 15] and asymmetric bellow actuator of about10mm diameter and 120mm length that was fabricated byrapid prototyping method to show large bending capabilities[16] are reported The above group [14 15] has not workedon the asymmetricity about the longitudinal direction of theactuatorsThe problem in the symmetric thickness actuator isthat it cannot withstand higher pressure ranges Also the rub-ber bonding process using ultraviolet light of 172 nm could bedangerous and expensive Our technique is further exploredand different arrangements of fibers are embedded within theactuator wall [17] The AFPAs show better deflection up tocertain amount of eccentricity provided in the asymmetricactuators The symmetric bellow actuators [3ndash6] do not pro-vide larger bending motion and do not withstand high pres-sures as compared to AFPA It is mentioned that in the case

Hindawi Publishing CorporationJournal of RoboticsVolume 2014 Article ID 902625 11 pageshttpdxdoiorg1011552014902625

2 Journal of Robotics

120∘

(a) (b) (c)

Figure 1 Cross section design of flexible pneumatic actuators (a) FMA (b) FPA and (c) AFPA

P

R

COPy

ex

M

Centroid

One convolution

Bellow side

Flat plate side

120579

rmrmsin120572

120572

L = 4na

Figure 2 Bending of asymmetric bellow actuator subjected to internal pressure

of symmetric actuators there is an unstable phenomenonthat occurs while gripping when the pressure of the workingfluid reaches some limit [3ndash6] In this paper an attempthas been made in the design and fabrication of miniaturesoft gripper based on the principle of AFPA The effect ofprofile shape and determining the optimum eccentricity toget the maximum deflection is analysed using Abaqus 613software The AFPAs are made almost semicircular in shapewith about 4mm radius and 30mm length The design ideaincludes the bellow structure on one side of the actuator andflat side on the other side The design shape also helps inholding the small objects on flat side of the actuator For thefabrication the actuator design is symmetrically divided intotwo equal halvesThen themanufacturing of each part is doneby compression molding process after making the requiredmold by micromachining process Then the two-half partsare bonded using single component room temperature vul-canising (RTV) silicone rubber Experiments are conductedwith the developed actuator by making a miniaturised softgripper for holding small parts AFPAs are developed as analternative to other actuation principles of today such aselectromotors shapememory alloysMcKibbenmuscles andflexible fluidic or pneumatic actuators These actuators havevarious advantages considering several criteria includingstress improvements packaging good power to weight ratioand high dynamicsThe cross section of AFPA is asymmetric

as compared to symmetric section of FMA or FPA as shownin Figure 1

2 Theory

The principle of working of asymmetric bellow actuatoris exactly opposite to that of the principle on which theBourdon tube is working The Bourdon tube used is initiallyin curved form with flat or elliptic cross section which underthe application of internal pressure will try to straighten upbecause of the action of the flat or elliptic section becomingcircular under pressure Contrary to this a straight asymmet-ric (eccentric) bellow tube with circular or semicircular crosssection under the application of pressure will become curvedand elliptic in cross section Asymmetric bellow actuator willbehave similar to the asymmetric tube actuators [8ndash10] butwill have higher flexibility and greater rate of expansion andcurving under internal pressure Bellows can be made asym-metric in cross section with either circular bellow or semicir-cular bellow Figure 2 shows bending of asymmetric flexiblesemicircular bellow actuator subjected to internal pressure

When fluid pressure 119875 is applied with the free end closedit bends due to combined effect of an end moment whichdevelops at the free end due to eccentricity and due todifferential expansion of the top and bottom fibers The forcegenerated due to pressure 119875 is given by 119865 = 119875 sdot 119860

119894 where 119860

119894

Journal of Robotics 3

Semicircular axis of asymmetric bellow

t

a

a

f

ro

ri

(a)

y

120579

120575max

120575b

Fb

rm

rmsin120572

y998400

Z998400

Z

(b)

Figure 3 Geometry description of (a) one bellow convolution (b) during bending

is the internal cavityrsquos area Since the cut-section is not axis-symmetrical its centroid is slightly shifted from the center ofpressure (COP) by a small distance ldquo119890rdquo Radius of curvatureat any pressure is 119877 = 119864119868119872 and the bending curvature isdefined as 119862 = 1119877

119872 is the moment acting at the free end 119864 is the youngrsquosmodulus and 119868 is the area moment of inertia

The moment due to eccentricity of the actuator is 119872119890=

119865 sdot 119890 where 119890 is the eccentricityThe deflection of the actuator is due to differential expan-

sion and the moment created due to the eccentricity of thegeometry [7ndash9] From the beam theory the vertical deflection1205750at the tip of the bellow and the angular deflection 120579

1are

given in terms of119872 as

1205750=1198721198712

3119864119868119910

(1)

1205791=119872119871

119864119868119910

(2)

where 119868119910is the area moment of inertia of the cross section of

asymmetric bellow and 119871 is the length of the bellowIn the case of differential expansion for the bellow it is

considered to have separate stiffness constant ldquo119896rdquo for the topand plate side of the bellow The force applied to the ABAis approximated to axial force and separated into 119865

119887and 119865

119901

respectively for both sides of the bellowThe total force is given by

119865 = 119865119887+ 119865119901= 119896119887120575119887+ 119896119901120575119901 (3)

where 120575119887and 120575

119901are the deflections at the top bellow side

and bottom flat part side respectively The 119896119887and 119896

119901are the

axial stiffness of the top bellow side and bottom flat part siderespectively

The axial stiffness 119896119887= 119861119868119910 is formulated using finite

element and linear regression analysis [18] withmodificationsfor semicircular asymmetric bellowThe simplified geometrydescription of one bellow convolution is shown in Figure 3

119861 is constant which depends on the geometry and thematerial of the bellow and is given by

119861 =

24119864 (4602 + 6 times 1071198863minus 862119903

0)

4119899 61205871198863 + 241198911198862 + 1198913 + 31198912119886120587 (1 + 1199052121198862) (4)

119868119910=120587 (1199030+ 119903119894) 1199053

12=1205871199031198981199053

6 (5)

where 119886 is radius of corrugation119891 is flank distance 119905 is bellowthickness 119903

0and 119903119894are outer and inner radius of the bellow 119903

119898

is the average radius of the bellow 119864 is modulus of elasticityand 119899 is number of convolutions of ABA

Thus deflection at bellow (top) side of the ABA is

120575119887=119865119887

119896119887

(6)

The deflection in the bottom plate part will be

120575119901=

119865119901

119896119901

(7)

Thus the angular deflection is given by

1205792=

120575119887minus 120575119901

119903119898

=

120575119887119901

119903119898

(8)

The total angular deflection of the asymmetric bellow actua-tor due to bending is obtained by adding (2) and (8) and isgiven by

120579 = 1205791+ 1205792=119872119871

119864119868119910

+

120575119887119901

119903119898

(9)

The moment due to differential expansion of the bellowand plate part of the bellow is assumed to be acting at thesemicircular axis of the asymmetric bellow section and themaximum deflection 120575max is at the top of the bellow


To determine the moment to bend the bellow due to dif-ferential expansion the deflection force which varies aroundthe circumference will be multiplied by the correspondinglevers and integrated around the circumference

From (6) 119865119887= 119896119887120575119887= 119861119868119910120575119887

Assuming an element force 119889119865 at any location at theaverage radius of the bellow is

119889119865 = 120575119887119861119868119910(119889120572

120587) (10)

For maximum deflection and substituting (5) into (10)

119889119865 = 120575max1198611199031198981199053

6sin120572119889120572 (11)

The element moment is

119889119872 = 119889119865119903119898sin120572 =

120575max11986111990531199032

119898

6sin2120572119889120572 (12)

The moment due to expansion of the bellow actuator is givenby

119872exp =120575max119861119905

31199032

119898

6int

120587

0

sin2120572119889120572 =120575max119861120587119905

31199032

119898

12=119865max1199031198982

(13)

where 119865max is the maximum force at the top of the bellowwhere it is subjected to maximum deflection

The total moment due to internal pressure 119875 is obtainedby

119872 = 119865 sdot 119890 +119865max1199031198982

(14)

3 Design and Analysis

The design of the AFPA is almost semicircular in shape suchthat on one half it has a bellow profile and on the other side itis flat This actuator is a single chambered tubular structurewhose thickness of one side is more than the thickness ofother side The actuator bends towards the thicker side dueto the differential expansion of top and bottom parts of theasymmetric bellow andmoment generated due to eccentricityof the bellowThe asymmetric bellow design gives maximumdeflection compared to normal symmetric designs up tocertain value of eccentricity and withstands high pressuresThe shape of the bellow profile also affects the deflectionof the actuator Figure 4 shows various bellow shapes Thepercentages of expansion when the internal pressure isincreased from 119875

0to 1198751are 26 47 48 and 70 for triangular

trapezoidal U shape and square shapes respectively Thesquare shaped bellow profile is more suitable as it givesmaximum deflection To decide on an optimum design forthe efficient actuation the FE analysis is done on Abaqus 613using models with different dimensions The material usedfor the actuator is a two-component silicone rubber which isRTV (room temperature vulcanising) type (KE-1606 Shin-Etsu Silicones Corp) Table 1 shows the material propertiesof the silicone rubber used in analysis and manufacturing

26

47

48

70

P0 P1

Figure 4 Effect of bellow shape on the expansion of actuator

Table 1 Material properties of silicone rubber

Properties Values1 Appearance Translucent2 Density 103 kgm3

3 Shore hardness 284 Tensile strength 43MPa5 Tear strength 12 kNm6 Elongation 350

Table 2 Parameters of AFPA

Model type 119860 [mm] 119861 [mm]Model 1 07 07Model 2 09 07Model 3 1 07Model 4 11 07Model 5 15 07

Figure 5 shows the model of the AFPA with dimensionscreated in the CAD software To find the optimum designdifferentmodels with varying parameters119860 and 119861 are createdfor analysis which are responsible for the eccentric actuation119860 and 119861 represent the thickness of flat plate side and thethickness of bellow side respectively As shown in Table 2model 1 to model 5 with different values of parameters areconsidered for analysis Length of the actuator in all modelsis 30mm and radius is about 3mm

Figure 6 shows the analysis results for the deflection ofAFPA The deflection is analysed using commercial finiteelement code Abaqus 613 Five models shown in Table 2 areused to measure the deflection with respect to the pressureapplied The asymmetric model 3 shows better deflectioncompared to the other models In model 1 the thicknessesof both flat plate side and bellow side are the same Model5 where the thickness of the flat plate side is twice thethickness of bellow side gives less deflection compared toall the other models Therefore it is observed that the deflec-tion of asymmetric bellow actuator is influenced by theeccentricity provided up to a certain extent This shows thatas the eccentricity increases by increasing the thickness ofthe flat plate side the stiffness also increases which reduces


30

mm

A

B

34mm

11mm

Figure 5 CAD model of the AFPA

0

10

20

30

40

50

60

0 50 100 150 200

Model 1Model 2Model 3

Model 4Model 5

Defl

ectio

n120579

(deg

)

Pressure P (kPa)

Figure 6 Deflection of AFPA for various models

the deflection of the actuator considerably In this papermodel 3 is chosen for manufacturing the actuator

Figure 7 shows the variation of bending angle withrespect to flat plate thickness (119860)Thebending angle increasesas the flat plate thickness increases up to 1mm After 1mm offlat plate thickness of the bellow part the bending angle startsdecreasing This means that there is an optimum bendingangle for certain amount of eccentricity of the actuator As theflat plate thickness of bellow actuator increases the stiffnessalso increases leading to decrease in bending angle

0

10

20

30

40

50

0 05 1 15 2Flat plate thickness (mm)

Bend

ing

angl

e120579

(deg

)

Figure 7 Variation of bending angle for different flat plate thickness(119860) of actuator

Figure 8 Fabricated convex and concave mold parts of AFPA

4 Manufacturing of AFPA

TheAFPA is manufactured using compressionmolding tech-nique For themanufacturing purpose the actuator is dividedinto two symmetrical equal halves along the axis Then eachhalf is manufactured separately and is bonded to make thecomplete AFPAThemanufacturing of each half part involvesmachining process for molds and preparation of the siliconerubber paste for molding Initially with appropriate dimen-sions from the simulations molds are manufactured usingmachining process which includes the convex and concavepart of the mold Figure 8 shows the mold manufacturedusing micromachining process This mold can be used tomanufacture half part of the actuator and similarly other twomolds are made for the other half part of AFPA

Now for the preparation of the silicone rubber pastewhose properties are as shown in Table 1 KE-1606 RTV sili-cone rubber is mixed with a 10 of curing agent (CAT-RG)Then this paste is poured between the convex and concavemold for both half parts of the actuator and then locked forcuring After curing both the parts are bonded using one-component RTV silicone rubberThe advantage in using one-component silicone rubber as bonding agent is that aftercuring its property of flexibility remains as compared to otheradhesive materials whichmakes the AFPA flexible as a singlecontinuous piece Figure 9 shows themanufactured half partsof the AFPA after molding process and the complete actuatorafter bonding the two half parts along with a pressure inlet

One-component Shin-Etsu KE 45 room temperature vul-canizing (RTV) silicone rubber compound is used for chem-ically bonding the two silicon bellow parts Shin-Etsu KE45RTV adhesive is a paste-like one-component material that


(a) (b) (c)

Figure 9 Manufactured parts of AFPA (a) outside view (b) inside view and (c) complete actuator

Camera

AFPA

Controller Relay Valves

Compressor Pressuresensor

Flow controller

Figure 10 Block diagram of experimental setup for the control of AFPA

cures when exposed to moisture in the air at room tempera-ture Due to its nonsag nonflowable features it may beapplied overhead or on side wall joints or surfacesThe bond-ing process involves first cleaning the rubber bellow partsand degreasing if necessary Apply the silicone rubber 1-component adhesive (KE 45 RTV) to a thickness of at least05mm Press together the bellow parts to be bonded and fixthem in position until the adhesive has curedThe curing timeis about 10 minutes Depending on the air humidity the opti-mum bond strength at room temperature is achieved within12 to 15 hours It is much faster at temperatures between 50and 100∘C and high humidity

5 Experimental Setup

Figure 10 shows the block diagram of experimental setuprequired for the control of the AFPA Figure 11 shows actualexperimental setup showing three AFPAs connected to airpressure supply via valves pressure sensor and flow con-troller The pressure inside the actuator is controlled by thecontrol of valves using a microcontroller The motion of theAFPA is captured by a camera

A 3-port solenoid valve having pressure input outputand exhaust ports is used in the experimental setup Input of

3 AFPAsPressure sensor

Solenoid valve

Figure 11 Experimental setup showing various devices

the valve is connected to the compressor output and outputof the valve is connected to the actuator input The principleof 3-port valve is to internally direct the pressure to the AFPAfrom the compressor and then to the exhaust from the AFPAcorresponding to the ON and OFF state of the valve Thecontrol of the valve is done by the Arm microcontroller viarelay circuit as the valve working voltage is 24V DC


Response time

03MPa P0 = 03MPa

P(t)

0MPa

500ms

SW ON

SW OFF

Figure 12 PWM control of the solenoid valve

Figure 12 shows the graphical relation for input and out-put of the solenoid valve When the valve is in ON state thepressure increases linearly in the AFPA Tomaintain the pres-sure inside the actuator at a constant value (eg 03MPa) for adesired position the valve should be controlled using a PWMpulse with 50 duty cycle as shown Stellaris LM3S808 Armmicrocontroller is used to generate PWM using 16-bit PWMmode

The design of the actuator is such that at a particularpressure the actuator reaches its maximum deflection afterwhich any increase in the pressure results in bulging of theactuator The time taken for the actuator to reach its maxi-mum pressure is called response time The controller isdesigned to maintain the maximum safe pressure in the actu-ator and also to maintain the position of the actuator at thedesired point To achieve this the ONOFF control of thesolenoid valves should be correspondingly done The config-uration of the valve is such that when the valve switch is ONthe pressure starts increasing and when its switch is OFF thepressure starts decreasing As shown in Figure 12 the upperpart shows the pressure in the actuator depending on theswitching state of the valve as shown in the lower part of thefigure Let the pressure 119875

0shown in Figure 12 be the required

pressure for the desired deflection or position of the actuatorTo maintain the pressure the valve should be switched onuntil the pressure inside the actuator reaches 119875

0after which

switching should follow a PWM with 50 duty cycle andthe frequency of 10Hz (maximum operating frequency ofthe solenoid valve) To generate the PWM and to read thepressure value from the pressure sensor Armmicrocontrolleris selected The controller is a 32-bit CPU with operatingfrequency of 50MHz which can generate five 16-bit PWMat a time Figure 13 shows the closed loop feedback controlsystem

The programming of the controller is carried out usingldquoCrdquo language Figure 14 shows the flow chart of controlprogram The 100 duty cycle implies the ON state of thevalve that results in the increase of pressure 50 duty cycleimplies the switching state of the valve that results in theconstant value of pressure and 0 duty cycle implies theOFFstate of the valve that results in the decrease of pressure

PWM generator

Air compressor

Pressure sensor

OPIP

Figure 13 Closed loop feedback control system

Start

Get pressure value

Compare with desired value

Constant P Decreasing PIncreasing P

Pd gt Pa

Pa

Pd lt Pa

Pd = Pa

Pd

Figure 14 Flow chart of control program

6 Results and Discussion

Figure 15 shows the deflection of manufactured AFPA withvarying pressures For every 30KPa increase in pressure thecorresponding deflection of the AFPA is shown The devel-oped actuator shows a very good position control with thePWM control technique And also the simple structure of theactuator with single chamber is more easy to be miniaturisedcompared to the conventional actuators of two or morechambers [3ndash6]

Figure 16 shows deflection analysis of the AFPA usingAbaqus 613 software Hyperelastic tetrahedron elements areused for rubber structures on wall of the actuator Appliedpneumatic pressures are given as incremental pressure loadin the software which always acts in the nominal directionon the rubber walls of the actuator In the FEM analysis theMooney-Rivlin model is used for approximating the char-acteristics of silicone rubber The coefficients are identifiedthrough the experimental results of plane strain tension testsof the silicone rubber

The deflection angles are 27∘ 30∘ 43∘ and 48∘ as againstthe 35∘ 37∘ 48∘ and 56∘ as obtained by the manufacturedactuator at internal pressures of 90 kPa 120 kPa 150 kPaand 180 kPa respectively There is a slight variation in theanalysis results as compared to experimental results Thiscould be due to error during modelling and analysis in thesoftware Also since the rubber material is elastic in natureand has highly nonlinear property it is difficult to analyzelarge deformations using software Due to pressure the rate atwhich the bending occurs is very high which again is difficultto predict theoretically with high accuracy


(a) 0 kPa

30∘

(b) 60 kPa

35∘

(c) 90 kPa

37∘

(d) 120 kPa

48∘

(e) 150 kPa

56∘

(f) 180 kPa

Figure 15 Deflection of the AFPA of model 3 at different pressures

27∘

(a)

30∘

(b)

43∘

(c)

48∘

(d)

Figure 16 Deflection analysis of model 3 AFPA for various internal pressures (a) 90 kPa (b) 120 kPa (c) 150 kPa and (d) 180 kPa

Figure 17 shows the static bending characteristic of thedeveloped actuatorThe analysis and experimental results areclose to each other The simple characteristic equations areobtained by assuming Youngrsquos modulus (119864) is constant andthe calculated characteristics are compared with experimen-tal data taken from a 6mm diameter

Figure 18 shows the path followed by the tip of the actu-ator at different pressures The pressure is varied from 0 kPato 180 kPa insteps of 30 kPa The displacement can be seen in

both 119909 and 119910 directions The displacement in both directionsincreases up to 150 kPa after which the tip of the actuatorstarts curling

The maximum force generated by the actuator is mea-sured by a load cell To detect the force one end of theactuator is fixed and the deflecting end is touching the loadcell The load cell is set up to measure the force from thetip of the actuator Figure 19 shows the experimental andtheoretical force curves of the actuator subjected to various


0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)

Experiment Experiment

Figure 17 The static bending characteristics of AFPA

0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)

Displacement in X (mm)

Trajectory of the actuator tipfor various pressures

Figure 18 Trajectory of the tip motion of the actuator

internal air pressuresThe pressure range is from 0 to 180 kPaThe force characteristics are almost linear and the maximummeasured force is 017N at 180 kPa

7 Miniature Soft Gripper

A miniature soft gripper consisting of three AFPAs is con-structed to pick and place small parts These three AFPAsare fixed to a silicone rubber plate at an angle of 120 degreesas shown in Figure 20 Figure 20 shows the various gesturesof the miniature soft gripper grasping IC chips The effect of

0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)

F (theory) (N)F (exp) (N)

Figure 19 Experimental and theoretical results of force measure-ment

instability while gripping when the pressure of the workingfluid reaches some limits as in the case of symmetric FPAs isavoided

In the three-chamber design of pneumatic actuators [3ndash5] it has been reported that when the pressure of the workingfluid reaches some limit an unstable phenomenon occursThe object and FMAs turn unstably around the polar axisof the actuator This is because of a net torque acting aboutthe polar axis due to the differential pressures in the variouschambers that causes twisting about the polar axis withpossible torsional buckling This problem is not observed inour design as there is only a single chamber and the pressurein this chamber does not create a net torque about the actua-torrsquos polar axis

8 Conclusion

In this paper a single chamber miniaturised asymmetricflexible pneumatic bellow actuator has been designed andfabricated which gives bending performance better than thesymmetric actuators of two or more chambers Using suchthree actuators a miniature soft gripper has also been devel-oped Analysis in ABAQUS software resulted in optimiseddesign of the actuator It is found that the effect of shapeand eccentricity of the AFPA plays an important role in thebending of the actuator and deflection of asymmetric bellowactuator is influenced by the eccentricity up to a certainextent The effect of instability while gripping when thepressure of the working fluid reaches some limits as in thecase of symmetric FPAs is avoidedThe assembled actuator inthe form of gripper has shown good results in picking of andplacing small parts It can also be shown that by supplyingvacuum or negative pressure to the actuator the actuator


(a) (b) (c)

(d)

Figure 20 Miniature soft grippers (a) without application of pressure (b) with application of pressure (c) grasping IC chip and (d) graspingup IC chip

can generate bending motion in counter direction of positivepressure These types of AFPAs will be useful in miniaturerobotic mechanisms where space is restricted or gentlehandling is required

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

The authors would like to thank KIST (Korea Institute ofScience and Technology) Korea for providing internship toAmrita University students

References

[1] L Zhang ZWang Q Yang G Bao and S Qian ldquoDevelopmentand simulation of ZJUT hand based on flexible pneumatic actu-ator FPArdquo inProceedings of the IEEE International Conference onMechatronics and Automation (ICMA rsquo09) pp 1634ndash1639 IEEEChangchun China August 2009

[2] I Gaiser S Schulz H Breitwieser and G BretthauerldquoEnhanced flexible fluidic actuators for biologically inspiredlightweight robots with inherent compliancerdquo in Proceedings of

the IEEE International Conference on Robotics and Biomimetics(ROBIO rsquo10) pp 1423ndash1428 Tianjin China December 2010

[3] K Suzumori S Iikura and H Tanaka ldquoApplying a flexiblemicroactuator to robotic mechanismsrdquo IEEE Control SystemsMagazine vol 12 no 1 pp 21ndash27 1992

[4] K Suzumori S Endo T Kanda N Kato and H SuzukildquoA bending pneumatic rubber actuator realizing soft-bodiedmanta swimming robotrdquo in Proceedings of the IEEE Interna-tional Conference on Robotics and Automation (ICRA rsquo07) pp4975ndash4980 Roma Italy April 2007

[5] K Suzumori S Iikura andH Tanaka ldquoDevelopment of flexiblemicroactuator and its applications to robotic mechanismsrdquo inProceedings of the IEEE International Conference onRobotics andAutomation pp 1622ndash1627 Sacramento Calif USA April 1991

[6] K Suzumori S Iikura and H Tanaka ldquoFlexible microactuatorfor miniature robotsrdquo in Proceedings of the IEEE Micro ElectroMechanical Systems Conference pp 204ndash209 Nora Japan 1991

[7] G Udupa Study and development of an unconventional devicefor industrial applications including robots and instrumentation[MS thesis] University BDT College of Engineering Davan-gere India 1992

[8] G Udupa and R Krishna Murthy ldquoA new flexing techniquefor soft Gripper designrdquo in Proceedings of the 16th All IndiaManufacturing Technology Design and Research Conference pp353ndash358 Bangalore India December 1994

[9] GUdupa P Sreedharan andKAditya ldquoRobotic gripper drivenby flexible microactuator based on an innovative techniquerdquoin Proceedings of the 6th IEEE Workshop on Advanced Robotics


and Its Social Impacts pp 1ndash6 Korean Institute of Science andTechnology Seoul Korea October 2010

[10] S Dinesh R Raveendran K Aditya P Sreedharan and GUdupa ldquoInnovativemicrowalking robot using flexiblemicroac-tuatorrdquo in Proceedings of the 28th International Symposium onAutomation and Robotics in Construction Seoul Republic ofKorea June 2011

[11] C P S Menon P Sredharan and G Udupa ldquoDesign and anal-ysis of multi-fingered dexterous hand based on an innovativeasymmetric flexible pneumatic actuatorrdquo in Proceedings of the2nd International Conference on Simulation Modeling andAnalysis pp 246ndash253 Vishwa Vidyapeetham CoimbatoreIndia 2011

[12] K B S Pavan Kumar S K Srinath C P Sankar Menon SPramod andGUdupa ldquoA novel technique for the developmentof an artificial human hand for prosthetic applicationrdquo inProceedings of the National Conference on Application of DataMining inManagement ofMetabolic andDegenerativeDisordersIndia pp 1ndash7 April 2012

[13] G Udupa ldquoArtificial robotic hand and process of manufactur-ing thereofrdquo Patent 3631CHE2011 2011

[14] S Wakimoto K Ogura K Suzumori and Y Nishioka ldquoMinia-ture soft hand with curling rubber pneumatic actuatorsrdquo inProceedings of the IEEE International Conference onRobotics andAutomation (ICRA rsquo09) pp 556ndash561 Kobe Japan May 2009

[15] S Wakimoto K Suzumori and K Ogura ldquoMiniature pneu-matic curling rubber actuator generating bidirectional motionwith one air-supply tuberdquo Advanced Robotics vol 25 no 9-10pp 1311ndash1330 2011

[16] Y Shapiro A Wolf and K Gabor ldquoBi-bellows pneumaticbending actuatorrdquo Sensors and Actuators A Physical vol 167no 2 pp 484ndash494 2011

[17] S Hirai T Masui and S Kawamura ldquoPrototyping pneumaticgroup actuators composed of multiple single-motion elastictubesrdquo in Proceedings of the IEEE International Conference onRobotics and Automation (ICRA rsquo01) vol 4 pp 3807ndash3812Seoul Republic of Korea May 2001

[18] M Hermann and A Jonsson Static characteristics of flexiblebellows [MS thesis] University of Karlskrona KarlskronaSweden 1997

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



120∘

(a) (b) (c)

Figure 1 Cross section design of flexible pneumatic actuators (a) FMA (b) FPA and (c) AFPA

P

R

COPy

ex

M

Centroid

One convolution

Bellow side

Flat plate side

120579

rmrmsin120572

120572

L = 4na

Figure 2 Bending of asymmetric bellow actuator subjected to internal pressure

of symmetric actuators there is an unstable phenomenonthat occurs while gripping when the pressure of the workingfluid reaches some limit [3ndash6] In this paper an attempthas been made in the design and fabrication of miniaturesoft gripper based on the principle of AFPA The effect ofprofile shape and determining the optimum eccentricity toget the maximum deflection is analysed using Abaqus 613software The AFPAs are made almost semicircular in shapewith about 4mm radius and 30mm length The design ideaincludes the bellow structure on one side of the actuator andflat side on the other side The design shape also helps inholding the small objects on flat side of the actuator For thefabrication the actuator design is symmetrically divided intotwo equal halvesThen themanufacturing of each part is doneby compression molding process after making the requiredmold by micromachining process Then the two-half partsare bonded using single component room temperature vul-canising (RTV) silicone rubber Experiments are conductedwith the developed actuator by making a miniaturised softgripper for holding small parts AFPAs are developed as analternative to other actuation principles of today such aselectromotors shapememory alloysMcKibbenmuscles andflexible fluidic or pneumatic actuators These actuators havevarious advantages considering several criteria includingstress improvements packaging good power to weight ratioand high dynamicsThe cross section of AFPA is asymmetric

as compared to symmetric section of FMA or FPA as shownin Figure 1

2 Theory

The principle of working of asymmetric bellow actuatoris exactly opposite to that of the principle on which theBourdon tube is working The Bourdon tube used is initiallyin curved form with flat or elliptic cross section which underthe application of internal pressure will try to straighten upbecause of the action of the flat or elliptic section becomingcircular under pressure Contrary to this a straight asymmet-ric (eccentric) bellow tube with circular or semicircular crosssection under the application of pressure will become curvedand elliptic in cross section Asymmetric bellow actuator willbehave similar to the asymmetric tube actuators [8ndash10] butwill have higher flexibility and greater rate of expansion andcurving under internal pressure Bellows can be made asym-metric in cross section with either circular bellow or semicir-cular bellow Figure 2 shows bending of asymmetric flexiblesemicircular bellow actuator subjected to internal pressure

When fluid pressure 119875 is applied with the free end closedit bends due to combined effect of an end moment whichdevelops at the free end due to eccentricity and due todifferential expansion of the top and bottom fibers The forcegenerated due to pressure 119875 is given by 119865 = 119875 sdot 119860

119894 where 119860

119894



t

a

a

f

ro

ri

(a)

y

120579

120575max

120575b

Fb

rm

rmsin120572

y998400

Z998400

Z

(b)







1are


1205750=1198721198712

3119864119868119910

(1)

1205791=119872119871

119864119868119910

(2)




119887and 119865

119901


119865 = 119865119887+ 119865119901= 119896119887120575119887+ 119896119901120575119901 (3)

where 120575119887and 120575



119901are the





119861 =

24119864 (4602 + 6 times 1071198863minus 862119903

0)

4119899 61205871198863 + 241198911198862 + 1198913 + 31198912119886120587 (1 + 1199052121198862) (4)

119868119910=120587 (1199030+ 119903119894) 1199053

12=1205871199031198981199053

6 (5)



119898



120575119887=119865119887

119896119887

(6)


120575119901=

119865119901

119896119901

(7)


1205792=

120575119887minus 120575119901

119903119898

=

120575119887119901

119903119898

(8)


120579 = 1205791+ 1205792=119872119871

119864119868119910

+

120575119887119901

119903119898

(9)




From (6) 119865119887= 119896119887120575119887= 119861119868119910120575119887


119889119865 = 120575119887119861119868119910(119889120572

120587) (10)


119889119865 = 120575max1198611199031198981199053

6sin120572119889120572 (11)


119889119872 = 119889119865119903119898sin120572 =

120575max11986111990531199032

119898

6sin2120572119889120572 (12)


119872exp =120575max119861119905

31199032

119898

6int

120587

0

sin2120572119889120572 =120575max119861120587119905

31199032

119898

12=119865max1199031198982

(13)



119872 = 119865 sdot 119890 +119865max1199031198982

(14)





26

47

48

70

P0 P1










30

mm

A

B

34mm

11mm


0

10

20

30

40

50

60

0 50 100 150 200


Model 4Model 5

Defl

ectio

n120579

(deg

)

Pressure P (kPa)




0

10

20

30

40

50


Bend

ing

angl

e120579

(deg

)








(a) (b) (c)


Camera

AFPA



Flow controller







Solenoid valve




Response time

03MPa P0 = 03MPa

P(t)

0MPa

500ms

SW ON

SW OFF






0after which



PWM generator

Air compressor

Pressure sensor

OPIP


Start

Get pressure value



Pd gt Pa

Pa

Pd lt Pa

Pd = Pa

Pd







(a) 0 kPa

30∘

(b) 60 kPa

35∘

(c) 90 kPa

37∘

(d) 120 kPa

48∘

(e) 150 kPa

56∘

(f) 180 kPa


27∘

(a)

30∘

(b)

43∘

(c)

48∘

(d)







0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)



0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)







0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)





8 Conclusion



(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











t

a

a

f

ro

ri

(a)

y

120579

120575max

120575b

Fb

rm

rmsin120572

y998400

Z998400

Z

(b)







1are


1205750=1198721198712

3119864119868119910

(1)

1205791=119872119871

119864119868119910

(2)




119887and 119865

119901


119865 = 119865119887+ 119865119901= 119896119887120575119887+ 119896119901120575119901 (3)

where 120575119887and 120575



119901are the





119861 =

24119864 (4602 + 6 times 1071198863minus 862119903

0)

4119899 61205871198863 + 241198911198862 + 1198913 + 31198912119886120587 (1 + 1199052121198862) (4)

119868119910=120587 (1199030+ 119903119894) 1199053

12=1205871199031198981199053

6 (5)



119898



120575119887=119865119887

119896119887

(6)


120575119901=

119865119901

119896119901

(7)


1205792=

120575119887minus 120575119901

119903119898

=

120575119887119901

119903119898

(8)


120579 = 1205791+ 1205792=119872119871

119864119868119910

+

120575119887119901

119903119898

(9)




From (6) 119865119887= 119896119887120575119887= 119861119868119910120575119887


119889119865 = 120575119887119861119868119910(119889120572

120587) (10)


119889119865 = 120575max1198611199031198981199053

6sin120572119889120572 (11)


119889119872 = 119889119865119903119898sin120572 =

120575max11986111990531199032

119898

6sin2120572119889120572 (12)


119872exp =120575max119861119905

31199032

119898

6int

120587

0

sin2120572119889120572 =120575max119861120587119905

31199032

119898

12=119865max1199031198982

(13)



119872 = 119865 sdot 119890 +119865max1199031198982

(14)





26

47

48

70

P0 P1










30

mm

A

B

34mm

11mm


0

10

20

30

40

50

60

0 50 100 150 200


Model 4Model 5

Defl

ectio

n120579

(deg

)

Pressure P (kPa)




0

10

20

30

40

50


Bend

ing

angl

e120579

(deg

)








(a) (b) (c)


Camera

AFPA



Flow controller







Solenoid valve




Response time

03MPa P0 = 03MPa

P(t)

0MPa

500ms

SW ON

SW OFF






0after which



PWM generator

Air compressor

Pressure sensor

OPIP


Start

Get pressure value



Pd gt Pa

Pa

Pd lt Pa

Pd = Pa

Pd







(a) 0 kPa

30∘

(b) 60 kPa

35∘

(c) 90 kPa

37∘

(d) 120 kPa

48∘

(e) 150 kPa

56∘

(f) 180 kPa


27∘

(a)

30∘

(b)

43∘

(c)

48∘

(d)







0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)



0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)







0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)





8 Conclusion



(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











From (6) 119865119887= 119896119887120575119887= 119861119868119910120575119887


119889119865 = 120575119887119861119868119910(119889120572

120587) (10)


119889119865 = 120575max1198611199031198981199053

6sin120572119889120572 (11)


119889119872 = 119889119865119903119898sin120572 =

120575max11986111990531199032

119898

6sin2120572119889120572 (12)


119872exp =120575max119861119905

31199032

119898

6int

120587

0

sin2120572119889120572 =120575max119861120587119905

31199032

119898

12=119865max1199031198982

(13)



119872 = 119865 sdot 119890 +119865max1199031198982

(14)





26

47

48

70

P0 P1










30

mm

A

B

34mm

11mm


0

10

20

30

40

50

60

0 50 100 150 200


Model 4Model 5

Defl

ectio

n120579

(deg

)

Pressure P (kPa)




0

10

20

30

40

50


Bend

ing

angl

e120579

(deg

)








(a) (b) (c)


Camera

AFPA



Flow controller







Solenoid valve




Response time

03MPa P0 = 03MPa

P(t)

0MPa

500ms

SW ON

SW OFF






0after which



PWM generator

Air compressor

Pressure sensor

OPIP


Start

Get pressure value



Pd gt Pa

Pa

Pd lt Pa

Pd = Pa

Pd







(a) 0 kPa

30∘

(b) 60 kPa

35∘

(c) 90 kPa

37∘

(d) 120 kPa

48∘

(e) 150 kPa

56∘

(f) 180 kPa


27∘

(a)

30∘

(b)

43∘

(c)

48∘

(d)







0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)



0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)







0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)





8 Conclusion



(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










30

mm

A

B

34mm

11mm


0

10

20

30

40

50

60

0 50 100 150 200


Model 4Model 5

Defl

ectio

n120579

(deg

)

Pressure P (kPa)




0

10

20

30

40

50


Bend

ing

angl

e120579

(deg

)








(a) (b) (c)


Camera

AFPA



Flow controller







Solenoid valve




Response time

03MPa P0 = 03MPa

P(t)

0MPa

500ms

SW ON

SW OFF






0after which



PWM generator

Air compressor

Pressure sensor

OPIP


Start

Get pressure value



Pd gt Pa

Pa

Pd lt Pa

Pd = Pa

Pd







(a) 0 kPa

30∘

(b) 60 kPa

35∘

(c) 90 kPa

37∘

(d) 120 kPa

48∘

(e) 150 kPa

56∘

(f) 180 kPa


27∘

(a)

30∘

(b)

43∘

(c)

48∘

(d)







0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)



0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)







0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)





8 Conclusion



(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










(a) (b) (c)


Camera

AFPA



Flow controller







Solenoid valve




Response time

03MPa P0 = 03MPa

P(t)

0MPa

500ms

SW ON

SW OFF






0after which



PWM generator

Air compressor

Pressure sensor

OPIP


Start

Get pressure value



Pd gt Pa

Pa

Pd lt Pa

Pd = Pa

Pd







(a) 0 kPa

30∘

(b) 60 kPa

35∘

(c) 90 kPa

37∘

(d) 120 kPa

48∘

(e) 150 kPa

56∘

(f) 180 kPa


27∘

(a)

30∘

(b)

43∘

(c)

48∘

(d)







0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)



0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)







0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)





8 Conclusion



(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Response time

03MPa P0 = 03MPa

P(t)

0MPa

500ms

SW ON

SW OFF






0after which



PWM generator

Air compressor

Pressure sensor

OPIP


Start

Get pressure value



Pd gt Pa

Pa

Pd lt Pa

Pd = Pa

Pd







(a) 0 kPa

30∘

(b) 60 kPa

35∘

(c) 90 kPa

37∘

(d) 120 kPa

48∘

(e) 150 kPa

56∘

(f) 180 kPa


27∘

(a)

30∘

(b)

43∘

(c)

48∘

(d)







0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)



0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)







0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)





8 Conclusion



(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










(a) 0 kPa

30∘

(b) 60 kPa

35∘

(c) 90 kPa

37∘

(d) 120 kPa

48∘

(e) 150 kPa

56∘

(f) 180 kPa


27∘

(a)

30∘

(b)

43∘

(c)

48∘

(d)







0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)



0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)







0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)





8 Conclusion



(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

10

20

30

40

50

60

0 50 100 150 200

TheoryFEM analysis

Pressure P (kPa)

Defl

ectio

n120579

(deg

)



0

5

10

15

20

25

30

35

0 5 10 15 20

Disp

lace

men

t in Y

(mm

)







0

002

004

006

008

01

012

014

016

018

0 30 60 90 120 150 180

Forc

e F

(N)

Pressure P (kPa)





8 Conclusion



(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










(a) (b) (c)

(d)





Acknowledgment


References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation






















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









research article asymmetric bellow flexible pneumatic...

Documents