The 2012 IEEE/ASME International Conference on
Advanced Intelligent Mechatronics
July 11-14, 2012, Kaohsiung, Taiwan
Nonlinear Mathematical Model of an Intelligent Pneumatic Actuator
(IPA) Systems: Position and Force Controls
Ahmad 'Athif Mohd Faudzi, Member, IEEE, Khairuddin Osman, M.F. Rahmat, Nu'man Din Mustafa,
M. Asyraf Azman, Koichi Suzumori, Member, IEEE
Abstract²This paper presents a nonlinear mathematical
modeling of an Intelligent Pneumatic Actuator (IPA) systems
and comparison of simulation results with existing experimental
results of position and force control data. The derivation method
of mathematical equation is referred and reviewed from several
literatures. From the mathematical modeling, a simulation
analysis was done and compared with existing data from
previous research. This paper proposed a reverse engineering
method from existing real system on the intelligent actuator that
focused more on development of hardware and experimental
setup to simulation analysis for validation of the developed
model. The simulation results show the affirmation of the
developed model based on the mathematical derivation work.
The simulation results demonstrate the open-loop and close-loop
control systems of position and force tracking control. Finally,
performance of this system are analyzed and compared. For
future research, this simulation can be used for development of
new controllers, validation process and can be applied to real
system.
I. INTRODUCTION
T his research focus on modeling and control of an
intelligent pneumatic actuator (IPA) systems. IPA is a
new developed actuator which integrates actuators,
microprocessors and various micro sensors. The advantages
of IPA are high power-to-weight ratio, lightweight,
comparatively low cost, easier maintenance, and have simpler
structure compared to other actuators. This intelligent
Manuscript received February 15, 2012. This work was supported in part
by the Universiti Teknologi Malaysia (UTM), Universiti Teknikal Malaysia
Melaka (UTeM) and Ministry of Higher Education (MOHE) Malaysia.
Ahmad 'Athif Mohd Faudzi is with the Department of Mechatronics and
Robotics Engineering, Faculty of Electrical Engineering, Universiti
Teknologi Malaysia, 81310 Skudai, Malaysia (corresponding author phone:
607-5535291; fax: 607-5566272; e-mail: athif@fke.utm.my).
Khairuddin Osman is with the Department of Industrial Electronics,
Faculty of Electrical and Electronics, Universiti Teknikal Malaysia Melaka,
Hang Tuah Jaya, 76100 Durian Tunggal, Melaka, Malaysia (e-mail:
khairuddin.osman@utem.edu.my).
M.F. Rahmat is with the Department of Control and Instrumentation
Engineering, Faculty of Electrical Engineering, Universiti Teknologi
Malaysia, 81310 Skudai, Malaysia (e-mail: fuaad@fke.utm.my).
Nu'man Din Mustafa is with the Department of Mechatronics and
Robotics Engineering, Faculty of Electrical Engineering, Universiti
Teknologi Malaysia, 81310 Skudai, Malaysia (e-mail:
nukmandin.88@gmail.com).
M. Asyraf Azman is with the Department of Mechatronics and Robotics
Engineering, Faculty of Electrical Engineering, Universiti Teknologi
Malaysia, 81310 Skudai, Malaysia (e-mail: masyrafazman@gmail.com).
Koichi Suzumori is with Graduate School of Natural Science and
Technology, Okayama University, Okayama, Japan. (e-mail:
suzumori@act.sys.okayama-u.ac.jp).
978-1-4673-2576-9/12/$31.00 ©2012 IEEE
pneumatic actuator was developed with new features which
will give better control, higher position and force accuracy,
communication ability and all-in-one mechanism for compact
system suitable for any related applications [1]-[3].
Intelligent actuator was initially developed by [4] where the
research concentrates on the development of the actuators and
highlighted new approach of designing servo systems. The
development comprises of control and fault tolerant software,
communication links, micro-controller, sensor system and
actuator. Results from a physical intelligent actuator was
given to show that the approach is feasible. [1] described the
actuator is very compact multisensory linear actuator and its
control system. The key element of this linear actuator system
is the DLR patented Planetary Roller Spindle Drive (PRSD).
With a volume of 50x50x104mm (length x width x height), it
can produced more than 100kg force and achieve a
displacement RI ´ with a resolution of 2µm.
Development of an intelligent pneumatic cylinder and its
application to pneumatic servo mechanism was proposed by
[5] that used micro functional elements such as
position/velocity/acceleration sensors, force sensors, micro
pneumatic control valves and a micro processor a built to
realize compact pneumatic servo systems with high control
performance. On the other hand, [6] developed a novel
cylinder by using various sensors and control devices. The
novel cylinders are an active icosahedron consisting of
intelligent cylinders and its application to virtual clay
modeling. Active link mechanisms are highly integrated and
enhanced by intelligent cylinders. A control system is built for
the active icosahedron. In the control system, the key element
is a control program implementing drawing of a virtual model
on display and controlling of active links. Virtual clays are
deformed by the program based on the apex positions
converted from cylinder lengths. The active icosahedron
realized dynamic interaction with virtual objects in PC,
showing the potential of the devices as a haptic interface.
This project also continued by [7] where intelligent
actuators were used to develop the active 80-faced polyhedron
for haptic physical human-machine interface. Intelligent
actuator is also applied for physical human-machine
interaction system which is Active Link Pillow (ALP)
developed by [8]. This research will discuss the enhancement
of previously developed pneumatic cylinder and proposed two
models of ALP as possible application of this actuator.
1105
 A modeling and characteristics analysis of intelligent
pneumatic muscle with Shape Memory Alloy (SMA) braided
shell was presented by [9]. This research is to realize the
pneumatic muscles mimic function of human muscles and
could be used as robot actuators.
Based on the research project from [5]-[7], a new intelligent
pneumatic actuator have been developed by [10]-[12] and
applied to Pneumatic Actuator Seating System (PASS) as an
application. Basically, the intelligent pneumatic actuator is
equipped with two sensors of optical encoder and pressure
sensor. A miniature valve is attached at the end of the cylinder
and a microcontroller board consists of Programmable System
on a Chip (PSoC) as the central processing unit is fixed at the
top of the actuator, in a single device. The intelligence aspect
of this actuator is that it can decide the output target based on
the feedback inputs with real-time communication capability.
The actuator has 200mm stroke and can give force up to lOON.
The 0.169mm laser stripe pitch can give high accuracy for
position control. An optical reflective surface mount encoder
chip is implemented on the bottom part of the PSoC circuit
board. This encoder chip consists of three parts; an LED light
source, a photo detector IC and optical lenses. The lenses
focus LED light onto the code strips on the guide rod and
reflected light on the photo detector IC. Fig. 1 shows overall
parts of the intelligent actuator. A model project is proposed
to be named as IPA to further improve the performance of the
current system in terms of applying new controllers, validation
process and apply to real system.
Fig. 1. Intelligent pneumatic actuator and its parts [11]
The nonlinear equation for the mathematical models from
previous research were used in this paper. The related
equations are used to model IPA and the model was tested
using experiment simulation model that allows the
measurement of the actuator's position and force outputs. The
simulation results demonstrate the open-loop and close-loop
control systems of position and force tracking control. The
experimental results will be compared and validate with the
existing results as [11]. Literature review about mathematical
modeling will be explained in detail in part III. The system are
1106
modeled in from three characteristics; Piston-Load Dynamics,
Model of the Cylinder Chambers and Valve Model.
II. IPA PRINCIPLES
Linear actuator type is a double acting (KOGANEI - HA
Twinport Cylinders) driving pneumatic actuators with two air
inlets and one exhaust outlet. Right and left movements of the
actuator can be controlled by manipulating the pressure in
chamber 1. The method of controlling the actuator movements
is by supplying constant air pressure to chamber 1 at 0.6 Mpa
(PI) while regulating air inside chamber 2 from (0-0.6) MPa
(P2). Right and left movements depend on the algorithm to
drive the valve using PSoC PWM duty cycle in chamber 1.
Pressure sensor is connected to PSoC for pressure data
reading. The chamber pressure is the input for the control
action of the cylinder. The pressure sensor reads the pressure
in chamber 2 and will be used to calculate force, F d using
equation (1) below:
(1)
where Pi and P 2 are pressure data, Ai and A 2 are
cross-sectional areas in chamber 1 and 2. Assume that Pi
(constant 0.6MPa), A], A 2 are known values. By reading the
pressure in chamber 2 (P 2 ) , force data, F d can be known.
The cylinder applies two valves (two ports two positions)
for driving the cylinder. The valves are attached at the end of
the cylinder as in Figure 1. The selection of these valves is
based on the function of the PASS [8], [10]-[12] where the
cylinder will not be driven in stable state (achieved target
position). This will reduce the valve operation where normal
on/off valve will always move to maintain certain position.
The method of controlling the valves is by using PWM duty
cycle driven by PSoC. Fig. 2 shows the IPA schematic
operations, valve connection and air flow to the cylinder.
Below are the possible movements of the cylinder which
depend on the valves operation.
1) Valve 1 off. Valve 2 off - cylinder stops
2) Valve 1 off. Valve 2 on - cylinder moves right direction
3) Valve Ion. Valve 2 off - cylinder moves left direction
4) Valve Ion. Valve 2 on - no operation
Parameters are taken based on data sheet and real IPA parts.
There was also a basic parameter taken from the other
references because no actual information obtained from IPA
parts. Table 1 shows the parameter constant as using in
simulation.
 Piston position, x external force, P1 and P2 are the absolute pressures in
Left direction Right direction DFWXDWRU¶V FKDPEHUV Pa is the absolute ambient pressure, A1
PSoC circuit and A2 are the piston effective areas, and Ar is the rod
board with
Cylinder encoder cross-sectional area. The right hand side of equation (3)
Chamber 1
A1, V1 A2, V2
Chamber 2 represents the actuator active force, produced by the pressure
Mp, Ar Load
Fd Laser strip code m F
differential acting across the piston. In order to control the
P1 P2
actuator force output, one has to fine tune the pressure levels
Ff
in the cylinder chambers using the pneumatic valve.
Pressure B. Model of Cylinder Chambers
sensor

Valve 2 Valve 1 The piston starting position is assumed to be at the center of

Fc2 Fc1 both chambers. Equation of this piston state have been
presented by [3], [13], [14], [17]. Therefore, the time
derivative for the pressure in the pneumatic cylinder chambers
xs2 xs1 becomes:
0.6MPa
RT P
Fig. 2. IPA schematic operations Pi k (mv ) k (r Ai x)
§1 · §1 ·
Voi Ai ¨ L r x ¸ Voi Ai ¨ L r x ¸
TABLE I ©2 ¹ ©2 ¹
PARAMETERS IPA
(4)
Symbol Quantity Values Remarks

d1 Actuator diameter 40 mm
where,
d2 Rod diameter 16 mm
L Rod stroke 200 mm mv (min mout ) (5)
Fmax Max force at 0.6 MPa 120 N
Fa Ambient Force 0N
k Specific heat ratio 1.4 ISO 6358 i=1 and 2, P is rate of change in pressure chamber, k is
T Temperature 294.5 K ISO 6358 specific heat ratio, Voi is inactive volume at the end of stroke
Pa Ambient pressure J/(kg.K) ISO 6358
R Gas constant K ISO 6358 and admission port, R is the ideal gas constant, P is pressure, T
mm = millimeter, N = Newton, K = Kelvin, J = joule, kg = kilogram.
is temperature, L is piston stroke, x is piston position, x is
rate of change piston position, A is are chamber area, min and
III. MATHEMATICAL MODEL mout are the mass flows entering and leaving the chambers.
The system dynamics equations are divided into three Equation (5) is the total mass flow rate for the valve. This
categories; Piston-Load Dynamics, Model of the Cylinder equation is important because mass flows entering and leaving
Chambers and Valve Model [2], [3], [13]-[20]. the chambers are controlled by two solenoid valves (on/off)
Basically, the dynamics of pneumatic cylinder considering when upstream and downstream pressures. [16] introduced
friction in chamber as in Fig. 2 can be described using the formula and present the upstream and downstream
1HZWRQ¶V 3ULQFLSOH pressures to valve are different for the charging and
discharging process of the cylinder chamber. The upstream
mx Fd F Ff (2) and downstream pressures are shown in equation (6) and (7).

where m, x , Fd, P1P2, Ff, F, A1A2 are mass, piston position, min i m( PS , Pi ) (6)
driving force, pressure, friction force, load force and cross
sectional area respectively. mout i m( Pi , Pa ) (7)
A. Piston-Load Dynamics
The equation of motion piston, rod and load can be expressed C. Valve Model
as combination equation (1) and (2);
Most of researcher considered valve as a critical component
of the actuator system [15]. The possible flow patterns in the
(M L M P ) x Ex F f FL P1 A1 P2 A2 Pa Ar (3) valve depends on the ratio of upstream to downstream
pressure as proposed by [19]. [17] present the advantages
where ML is the external load mass, MP is the piston and rod on/off valve design which is simple, have flexible ports
assembly mass, x is the piston position, is the viscous design, small internal leakage, fast response, and small
friction coefficient, Ff is the Coulomb friction force, FL is the

1107
internal friction. The standard equation for the mass flow V. RESULT AND ANALYSIS
through an orifice of area Av used by [15], [17] is: Towards realizing the operation of IPA, we have developed
open-loop and close-loop control system for position and
- Pu Pd force tracking control. Open-loop control system is a control
°C f Av C1 if d Pcr
T Pu
° system that does not have a feedback loop thus does not have
mv m( Pu , Pd ) ® 1/ k ( k 1) / k
°C A C Pu § Pd · §P · Pd self-correcting function. While for close-loop control system,
¨¨ ¸¸ 1 ¨¨ d ¸¸ if ! Pcr
° f v 2 it is a control system with a feedback loop for fixing the errors.
¯ T © Pu ¹ © Pu ¹ Pu
Previous work done by [11] presented experimental
(8)
evaluation of Proportional Integral (PI) controller for position
and force tracking control using close-loop control system.
where equation (6) and (7) insert to equation (8) and
The effects of proportional gain (Kp) and integral gain (Ki) in
the close-loop control system will stabilize the system in
2 §S · (9)
AV X spool *¨ ¸, X spool CV u eliminating errors. However, results in [11] do not have
©4¹ comparison model for validation. Therefore, this research
tries to model the IPA using open-loop control system and test
k 1/ k 1 k /k 1
k§ 2 · 2k § 2 · the close-loop control system using the same controller and
C1 ¨ ¸ , C2 , Pcr ¨ ¸ input-output response.
R © k 1¹ R(k 1) © k 1¹
(10) A. Open-loop control system
The IPA system model was developed using mathematical
mv is the flow through the valve orifice, Cf the coefficient of equations model in Section III. In order to ensure that the
flow, Pu the upstream pressure, Pd the downstream pressure, output response will follow the input signal, IPA system is
Pcr is critical pressure ratio, Xspool is spool displacement, CV is initially tested using open-loop control system. By
valve constant and u is voltage input. implementing the model into MATLAB Simulink, the result
of multi sine response (sample time = 0.1s) shown in Fig. 5
IV. DEVELOPMENT OF THE IPA SIMULATION MODEL and evaluated system by frequency response shown in Fig. 6.
For the design and analysis of an automatic control system, It can be noticed that the IPA open-loop control system
it is important to create a dynamic mathematical model. The response gives big error from the input signal as no controller
complete mathematical model of this intelligent pneumatic is being applied. However, frequency response shown in Fig.
actuator system consists of the valve dynamics equation, two 6 is acceptable to be used for future IPA system .
equations for chamber pressure time derivatives and the
1.5
piston-load equation of motion. The whole equation is related Input
to one another. 1
From the model, controller design can be simulated before 0.5
Position (m)

real-time experimental evaluation. In this research we are

0
using MATLAB Simulink environment. Figure 3 and 4 shows
the schematic simulation models for open-loop and close-loop -0.5
of IPA system respectively. The IPA characteristic dynamics -1
are presented by four subsystems which are valve, mass flow
-1.5
rate, chambers and piston. 0 10 20 30 40 50 60 70 80 90 100
Time (s)
(a) Input response
Intelligent Pneumatic Actuator (IPA)

Chamber 1 0.01
Output
+
m(Ps , P1 ) m1 P1 0.005
Valve 1 -
ui
(Inlet) Av m(P1 , Pa ) Position (x)
Position (m)

Valve 2 Xspool
P
x Force (F)
uo + 0
(Outlet) m(Ps , P2 ) m2
Input Output
Valve
- P2 Piston
m(P2 , Pa )
-0.005
Mass Flow Rate Chamber 2

Fig. 3. Schematic simulation model (Open-loop)

-0.01
0 10 20 30 40 50 60 70 80 90 100
Time (s)
+ IPA
u Controller
(Plant)
x, F (b) Output response
- Fig. 5. Open-loop multi sine response

Fig. 4. Schematic simulation model (Close-loop)

1108
 Frequency response
TABLE II
Amplitude
10
0 SPECIFICATIONS CONTROLLER RESPONSES FOR POSITION TRACKING
Controller
Specifications
-10
PI
10
10
-10
10
-5
10
0
10
5 Percent Overshoot (%OS) 35.3%
0 Peak Time (TP) 0.2083s
Settling Time (TS) 5.8s
Phase (deg)

-200 Rise Time (TR) 0.109s

Percent Steady State error (%ess) 0.01%
-400 -10 -5 0 5 % = percent, s = second.
10 10 Frequency (rad/s) 10 10

Fig. 6. Bode diagrams

The result for PI controller for force tracking control is
B. Close-loop control system shown in Fig. 10. The output force of IPA follows the input
The result of PI controller for position tracking control is given. However, some error is observed due to the mechanical
shown in Fig. 7-9 with different input signal. From the results, friction from certain inaccurate parameters in the model.
close-loop control system give better response compared to
open-loop control system. The PI controller gives a better 120
Output
steady state response. Table II shows the controller responses 100 Input
Reference
for step input position tracking in Fig. 7. The overshoot 80

percentage, %OS is 35.3%, peak time, TP is 0.2083s, settling

Force (N)
60
time, TS is 5.8s, rise time, TR is 0.109s and percent steady state 40
error, (%ess) is 0.01%. The steady state response is improving 20
about 99.78% comparing to the open-loop control system. 0

0.14 -20
0 20 40 60 80 100 120
Output Time (s)
0.12 Input
Reference
Fig. 10. Close-loop multi step response for force tracking using PI control
0.1
Position (s)

0.08 VI. DISCUSSION

0.06 From the simulation result, the experimental test of position
0.04 and force control in [11] can be validated. However during the
0.02 mathematical model development, there are certain
0
assumption is being used. The valve control in simulation is
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s) different from the real experiment. In experimental evaluation
Fig. 7. Close-loop step response for position tracking using PI control in [11], valve control was focused only on Chamber 1 (as
0.15 explained in IPA principles). However, this method cannot be
Output
0.1 Input
shown in the simulation work because the basic reference of
Reference air inlet and outlet of the valve need both chambers.
0.05
Nevertheless, the results obtained from the simulation are
Position (m)

0
similar with experimental results. Percentage overshoot of
-0.05
simulation development higher than real development results
-0.1
but still accordance with tracking input response. Table III
-0.15 shows the differences between hardware and simulation
-0.2
0 20 40 60 80 100 120 analysis results. There are significant differences in the
Time (s)
controller, overshoot response and contraction movement
Fig. 8. Close-loop sine wave response for position tracking using PI control
analysis. The value of the gain Kp and Ki in simulation are
0.12 Output larger than the experimental work. These changes affect the
Input
0.1
Reference overshoot response where simulation results are higher than
0.08 the experimental results. However, contraction movement to
Position (m)

0.06 hardware have delay because of real time situation,

0.04
displacement and compression problem.
0.02

-0.02
0 20 40 60 80 100 120
Time (s)
Fig. 9. Close-loop multi step response for position tracking using PI control

1109
 TABLE III
DIFFERENCES BETWEEN SIMULATION AND HARDWARE ANALYSIS
RESULTS
Analysis Hardware [11] Simulation
Controller Value Kp and Ki Value Kp and Ki
small big
Overshoot Response Low High
Contraction movement Delay No delay
Kp = Proportional gain, Ki = Integral gain.
VII. CONCLUSION
In modern era, pneumatic system is getting more
complex because of the evolution in mechatronics systems,
where integration of mechanical, electrical and electronics,
control systems and modern control algorithms becomes more
active. This research tries to combine all components by
proposing a new design concept of pneumatic actuator called
IPA by integrating several sensors and control algorithms in
one unit. From the mathematical model and simulation results
in open-loop and closed-loop control systems, it can be
validated with the previous experimental work in [11]. This
research will provide greater opportunities for future work
such as development of new controllers, validation process
and can be apply to real system.
ACKNOWLEDGMENT
The authors would like to thank Universiti Teknologi
Malaysia (UTM) under UTM-NAS Grant No.
R.J130000.7723.4P008, Universiti Teknikal Malaysia
Melaka (UTeM), Ministry of Higher Education (MOHE)
Malaysia and Okayama University for their support.
REFERENCES
[1] Y. Boulenger, E. Kramer, L. Hong, N. Seitz and G. Hirzinger, "An
intelligent linear actuator and its control system," in Advanced
Intelligent Mechatronics, 2001. Proceedings. 2001 IEEE/ASME
International Conference on, 2001, pp. 521-526 vol.1.
[2] H. I. Ali., S. B. B. M. Noor., S. M. Bashi. and M. H. Marhaban., "A
Review of Pneumatic Actuators (Modeling and Control)," Australian
Journal of Basic and Applied Sciences, vol. 3(2), pp. 440-454, 2009.
[3] H. I. Ali., S. B. B. M. Noor., S. M. Bashi. and M. H. Marhaban.,
"Mathematical and Intelligent Modeling of Electropneumatic Servo
Actuator Systems," Australian Journal of Basic and Applied Sciences,
vol. 3(4), pp. 3662-3670, 2009.
[4] P. Strickland, J. E. L. Hollis, J. Billingsley and B. Haynes, "Intelligent
Actuators," in Intelligent Control and Instrumentation, 1992. SICICI
'92. Proceedings., Singapore International Conference on, 1992, pp.
630-635.
[5] K. Suzumori, J. Tanaka and T. Kanda, "Development of an intelligent
pneumatic cylinder and its application to pneumatic servo
mechanism," in Advanced Intelligent Mechatronics. Proceedings,
2005 IEEE/ASME International Conference on, 2005, pp. 479-484.
[6] J. Ochi, T. Hashimoto, J. Tanaka, K. Suzumori and T. Kanda,
"Development of active icosahedron and its application to virtual clay
modeling," in Micro-Nanomechatronics and Human Science, 2004
and The Fourth Symposium Micro-Nanomechatronics for
Information-Based Society, 2004. Proceedings of the 2004
International Symposium on, 2004, pp. 175-180.
[7] T. Kubo, Y. Kobayashi, A. A. Mohd Faudzi and K. Suzumori,
"Development of Active 80-faced Polyhedron for haptic physical
human-machine interface," in Intelligent Robots and Systems, 2009.
1110
IROS 2009. IEEE/RSJ International Conference on, 2009, pp.
1259-1264.
[8] Ahmad 'Athif Mohd Faudzi, "Development of Intelligent Pneumatic
Actuators and Their Applications to Physical Human-Mechine
Interaction System," Ph.D. thesis, The Graduate School of Natural
Science and Technology, Okayama University, Japan, , vol.
September, 2010.
[9] W. Binrui, Z. Weiyi, J. Yinglian, X. Hong and K. Haisen, "Modeling
and Characteristics Analysis of Intelligent Pneumatic Muscle with
Shape Memory Alloy Braided Shell," in Informatics in Control,
Automation and Robotics, 2009. CAR '09. International Asia
Conference on, 2009, pp. 3-7.
[10] A. A. M. Faudzi, K. Suzumori and S. Wakimoto, "Distributed Physical
Human Machine Interaction Using Intelligent Pneumatic Cylinders," in
Micro-NanoMechatronics and Human Science, 2008. MHS 2008.
International Symposium on, 2008, pp. 249-254.
[11] A. A. M. Faudzi, K. Suzumori and S. Wakimoto, "Design and control
of new intelligent pneumatic cylinder for intelligent chair tool
application," in Advanced Intelligent Mechatronics, 2009. AIM 2009.
IEEE/ASME International Conference on, 2009, pp. 1909-1914.
[12] A. A. Mohd Faudzi, K. Suzumori and S. Wakimoto, "Development of
Pneumatic Actuated Seating System to aid chair design," in Advanced
Intelligent Mechatronics (AIM), 2010 IEEE/ASME International
Conference on, 2010, pp. 1035-1040.
[13] E. Richer and Y. Hurmuzlu, "A High Performance Pneumatic Force
Actuator System: Part I---Nonlinear Mathematical Model," Journal of
Dynamic Systems, Measurement, and Control, vol. 122, pp. 416-425,
2000.
[14] E. Richer and Y. Hurmuzlu, "A High Performance Pneumatic Force
Actuator System: Part II---Nonlinear Controller Design," Journal of
Dynamic Systems, Measurement, and Control, vol. 122, pp. 426-434,
2000.
[15] C. Ying, Z. Jia-fan, Y. Can-jun and N. Bin, "Design and hybrid control
of the pneumatic force-feedback systems for Arm-Exoskeleton by
using on/off valve," Mechatronics, vol. 17, pp. 325-335, 2007.
[16] M.F. Rahmat, Sy Najib Sy Salim, $KPDG µ$WKLI 0RKG )DXG]L =RRO
Hilmi Ismail, S.I.Samsudin, N.H.Sunar and K.Jusoff, "Non-linear
Modeling and Cascade Control of an Industrial Pneumatic Actuator
System," Australian Journal of Basic and Applied Sciences, vol. 5, pp.
465-477, 2011.
[17] V. %ODJRMHYLü DQG 0 6WRMLOMNRYLü 0DWKHPDWLFDO DQG 6LPXOLQN
Model of the Pneumatic System With Bridging of the Dual Action
Cylinder Chambers," Facta univesitaties series Mechanical
Engineering, vol. 5(1), pp. 23-31, 2007.
[18] Z. Yong and E. J. Barth, "Impedance Control of a Pneumatic Actuator
for Contact Tasks," in Robotics and Automation, 2005. ICRA 2005.
Proceedings of the 2005 IEEE International Conference on, 2005, pp.
987-992.
[19] T. JM., C. T., K. H. and L. M., "Dynamic behavior of pneumatic
systems for lower extremity extenders," IEEE Int Conf Robotics
Automation, pp. 2348±53, 2002.
[20] Igor L. Krivts, German V. Krejnin ³3QHXPDWLF $FWXDWLQJ 6\VWHPV IRU
$XWRPDWLF (TXLSPHQW 6WUXFWXUH DQG 'HVLJQ ´ United States: Taylor
& Francis, 2006, pp. 56±69.