The braided pneumatics actuator was invented by McKibben to help the movement
of polio patients in 1950. They are also called McKibben muscles or artificial
muscles powered by the compressed air (Laksanacharoen, 2004).
It was in 1958 that R.H Gaylord discovered a pneumatics actuator which was used
as the original applications such as, a door opening arrangement and an industrial
hoist. These muscles were made from pure rubber latex, covered by a double helical
braided wire which would contract when expanded radially. A Pneumatic Artificial
Muscle (PAM) is a pneumatic actuator for converting pneumatic power to pulling
force (Wickramatunge and Leephakpreeda, 2009).
In 1980s, the PAMs were designed and more powerful PAMs were introduced by
Bridgestone Company (Wickramatunge and Leephakpreeda, 2009).
The PAMs were used for painting applications and for assist disabled individuals
and service robotics. Conventional Pneumatics actuators with piston and cylinder
were widely used in factory automation. A muscle produces strength ten times
greater than the strength of a conventional pneumatics actuator (Wszolek
et al., 2007).
Lately robotic systems have started using pneumatics as a main motion power source. The main reasons for using pneumatics are: easy availability, dependability, flexibility, cleanliness, surges of power, linear motions, variable speed and safety. Apart from the above advantages there are a few limitations as mentioned below: low thrust, non-availability servo action. Several types of pneumatics actuators like cylinders, pneumatics engines and even pneumatics stepper motors are commonly used now days. A less well known type is Pneumatics Artificial Muscles (PAMs).
Characteristics of pams: PAMs are operated by pneumatic pressure and
are contractible naturally upon inflation. Their construction materials simply
consist of flexible inflatable membrane, reinforced with fibrous filament and
fitted with gas closure fittings for mechanical load-carrying at its end. As
the membrane is pressurized, it bulges outward in radial direction, while contracting
in length along is axial direction. It is this axial contraction where PAM exerts
a pulling force on its end-effectors. This force generated from contraction
and the subsequent motion on the loads is unidirectional (Ramasamy
et al., 2005). The most commonly used design to date, the McKibben
(Lynn, 2008) muscle, has some drawbacks, mainly with regard
to its control but also with regard to service life: the flexible membrane connected
to rigid end fittings which introduces stress concentration and there from possible
membrane ruptures. Cylinder, being entirely composed of rigid materials, does
not suffer from these problems.
Application in robotics: Wickramatunge and Leephakpreeda
(2009) reported that the low assembly weight and the high power-to-weight
ratio of PAMs are most desirable to be considered for use in mobile robotics.
In certain situation, the impotency of the PAMs is higher than the conventional
The development empirical model gives a concrete and effective description to understand the mechanical behavior of the PAMs for design and usage. The stiffness parameter of the muscle denotes as K and it is considered a function of the operating air pressure, P (Gauge pressure) and stretched length, Ls. The elastic forces adversely generated by the PAMs is denoted by Felastic and the expression given in Eq. 1 shows the proposed model for the force acting on the PAMs as a function of K and Ls:
In this study, the stiffness parameter K is taken as a second order polynomial
function of Ls and P and it is expressed in Eq. 2.
However, a universal approximation method with fuzzy logic or neural network
can be used as alternative choice for accuracy viewpoint (Vanderborght,
where ao, a1, a2, a3 are constant
coefficients, which can be obtained from experiment data. Typical values of
coefficients in K for operating on construction mode under low pressure:
a0 = -2.977; a1 = -5.890; a2=
3.203 and a3 = 104.824
In the robotic application, one has to find suitable actuators which can satisfy
the need for the strong and lightweight actuators. PAMs could very well suit
in the near future for such needs. Pleated muscles have been proven to be able
to perform very accurate positioning tasks employing PAMs (Vanderborght,
Ramasamy et al. (2005) reported the properties
influencing the PAMs for their load-carrying capacity and low weight capability
in assembly. Discussion on their designs and capacity to function as locomotion
device in robotics applications have been analyzed by Finite Element (FE) modeling,
to represent PAMs overall structural behavior under any potential operational
conditions. He also (Ramasamy et al., 2005a)
reported the construction of PAMs mainly consisting of flexible, inflatable
membranes, having orthotropic materials behavior. In this study, the properties
of the PAMs are explained in terms of their load carrying capacity and low weight
Vanderborght (2006) is describing the implementation
of Pleaded Pneumatic Artificial Muscles (PPAM) into innovative robotic application.
This actuator has a very high power to weight ratio. Nowadays legged robots
are gaining more and more interest. But most of the robots are using electrical
drives, making these machines heavy and power consuming. An actuator, such as
the PAM lowers the robot weight and the artificial muscle can be exploited to
reduce energy consumption.
Deaconescu 2007 presented some of the result of research
carried out in the Fluidtronics Laboratory of the Regional festo training centre.
The experiments aimed at determining the variation diagrams of the feed pressure
and the flow of consumed air for a complete cycle of the analyzed pneumatic
muscle. Further, the response times of the muscle could be determined for various
values of the feed pressure and the influence of adding a quick exhaust valve
to the circuit on the response time for muscle deflation could be studied. Figure
1 presents the several constructive forms of pneumatic muscles and Fig.
2 presents the evolution in time of constructive solutions for pneumatic
Kawashima et al. (2004) has described the usage
of Fiber Knitted Pneumatic Artificial muscle Rubber Muscle (PARM) which was
selected as the actuator for the arm. This arm can be designed for easy installation
and portability. Experimental results conducted by the setup shown in the Fig.
3, showed that the developed system successfully operated in material handling
system by remote operation.
Recently robotics field has been a choice for this PAM actuator usage because
of its lightweight and high strength. Pneumatics Artificial Muscle seems a better
choice than present day electric or other device (Daerden,
and Lefeber, 2002). The pleated muscles have been proven to be able to perform
very accurate positioning tasks. Using PAMs, the structure of a robot arm can
be made a lot lighter and its payload to weight ratio a lot higher compared
to electric drives.
Especially for mobile robot demanding lightweight actuators will be able to
generate high torques at low and moderate speeds, Robotic arms can be connected
to the structure without having to use gearing and they have better shock resistance
capability and can be employed easily for autonomous operation (Daerden
and Lefeber, 2002). In 2002, Daerden and Lefeber (2002)
reported although PAMs has been around for quite time, this actuator has not
been used. The author attributed the following reasons: the lack of large scale
need for this specific type of actuator and the lack of technological effort
to improve the existing designs.
Ahn and Nguyen, 2007 presented the solution for position
control of a robot arm with a slow motion driven by two pneumatic artificial
muscles as depicted in Fig. 4.
In this study, the author explains some limitations such as a deterioration
of the performance of transient response due to change in the external load.
To overcome this problem, a switching algorithm of the control parameter using
a learning vector quantization neural network (LVQNN) is proposed.
Wang et al. (2008) proposed a novel humanoid
robot eye, which is driven by six Pneumatic Artificial Muscles (PAMs) and rotates
with 3 Degree Of Freedom (DOF) as in Fig. 5. The design of
the mechanism and motion type of the robot eye are inspired by that of human
eyes. The model of humanoid robot eye is established as a parallel mechanism
and the inverse-kinematic problem of this flexible tendons driving parallel
system is solved by the analytical geometry method. As an extension, the simulation
result for saccadic movement is presented under three conditions.
The design and kinematic analysis of the prototype could be a significant step
towards the goal of building an autonomous humanoid robot eye with the movement
and especially the visual functions similar to that of human.
Performance evaluation of pams: In Deaconescu and
Deaconescu, (2008) presented results of the ongoing theoretical and experimental
research on the operational behavior of pneumatic muscle. Information is presented
concerning muscle structure, as well as data on the generated forces, positioning
accuracy and the maximum working frequency of these pneumatics actuation elements.
Figure 6 elaborate the stroke carried out by a pneumatic
muscle depending on the level of the feed pressure. Pneumatic muscles are actuating
elements that transform pneumatic energy into mechanical energy. A characterization
of the behavior and the performances of these actuating elements entail both
analytical calculations and thorough experimental research. In order to determine
the functional characteristics of a pneumatic muscle, first its structure needs
to be explored. A pneumatic muscle includes an interior tube of various length
made from an elastic material, typically neoprene. This tube is wrapped in a
multi-layer nylon tissue ensuring its strength and protection from the influences
of the working environment. Figure 7 shows the geometry of
the wrapping envelope Deaconescu and Deaconescu (2008).
In Lynn (2008) described the use of biomechanical phenomenological
model to a commercially available pneumatic muscle actuator. Experiment has
been conducted that defines boundaries of operation where linear approximations
can be used to describe the dynamics of PAM. The dissertation shows that nonlinearities
exist more prevalently at higher loads in the PAMs.
In Toman et al. (2009) designed an apparatus
which would enable experimental investigation of PAMs. The stand described in
this study is a didactic laboratory stand, which can investigate and gather
knowledge of construction of working elements such as: a fluidic muscle, a PLC
controller, DSP system as well as proportional pressure control technique. The
stand was designed and visualized by utilization of professional CAD software-Autodesk
In Kuriyama et al. (2009) proposed a method
of estimating the length from the circumferential displacement, which can be
measured by a sensor made of electro conductive, flexible rubber. Higher accuracy
is obtained by measuring the circumferential displacement than by measuring
the axial displacement using the sensor. The flexibility of the sensor enables
us to accurately control the actuator without any loss of flexibility or increase
in weight. Furthermore, the sensor does not require the attachment of any rigid
fixtures. The accuracy of the estimate is successfully evaluated and the usefulness
of the proposed method is verified through its application to a multi-link arm.
In Shen (2010) proposed a control methodology of the
pneumatic artificial muscle (PAM) which actuated two servo systems. The four
major processes including the flow dynamics, pressure dynamics, force dynamics
and a load dynamics are studied to develop a full non linear model taking the
valve command as the input and calculating the third order derivative of the
load position as the output. Based on this model, the standard sliding mode
control approach was applied to obtain robust control in the existence.
Chang (2010) developed an adaptive self-organizing
fuzzy sliding mode control (ASOFSMC). Its fuzzy sliding surface can help reduce
the number of fuzzy rule. The self-organizing mechanism is employed to modify
fuzzy rules online. The model matching technique is then adopted to adjust the
scaling factors. In Wickramatunge and Leephakpreeda, 2010
proposed models that are experimentally derived to describe mechanical behaviors
of the PAMs. The experimental results show a non-linear relationship between
contraction as well as air pressure within the PAMs and a pulling force of the
PAMs. Three different sizes of PAMs available in industry are studied for empirical
modeling and simulation. The case studies are presented to verify close agreement
on the simulated results to the experimental results when the PAMs perform under
In Zhu et al. (2008) described a discontinuous
projection-based adaptive robust control strategy is being adopted to compensate
for both the parametric uncertainties and uncertain nonlinearities of a three-pneumatic-muscles-driven
parallel manipulator to achieve precise posture trajectory tracking control.
The resulting controller effectively handles the effects of various parameter
variations and the hard-to-model nonlinearities such as the friction forces
of the pneumatic muscles. Simulation and experimental results are obtained to
illustrate the effectiveness of the proposed adaptive robust controller.
Zhang et al. (2008) proposed a novel curved
pneumatic muscle based rotary actuator for the wearable elbow exoskeleton with
joint torque control.
Compared to the general utilization of the Pneumatic Muscle Actuator (PMA)
in a rotary joint, this novel structure weakens coupling relationship between
the output torque/force and contacting displacement of the PMA so that it can
be easily utilized in the tele-robotics with torque/force-feedback or the exciting
application in rehabilitation for a wide range. By referred to two physical
models, namely beam model and membrane model, the mechanics properties of this
mechanical structure is analyzed.
||Scheme of the novel curved PMA rotary actuator in the wearable
elbow exoskeleton (Zhang et al., 2008)
In addition a hybrid fuzzy controller composed of bang-bang controller and
fuzzy controller is employed for output torque control with high accuracy as
well as fast response. In a series of experiments, the actuator exhibits both
good static and dynamic performances that well validated the models and control
strategy. Figure 8 shows scheme of the novel Curved PMA rotary
actuator in the wearable elbow exoskeleton.
In Zhang et al. (2005) reviews the working principles
of pneumatic muscle actuators, on the application in parachute systems for soft-landing
and steering control. A new finite element model for pneumatic muscle actuators
is presented. Geometrically nonlinear anisotropic membrane elements are used
in this model to simulate the nonlinear structural dynamic behavior of pneumatic
muscle actuators, which is different from previous approaches. A quasi-static
pneumatic muscle actuator model is analyzed for validation and two dynamic applications
of pneumatic muscle actuators in parachute systems are also presented.
Zhou et al. (2004) stated a special cable element
which is developed to model the mechanical behavior of PMAs. The new element
provides relationships between the PMA internal pressure, fiber bias angle,
PMA radius and length, and resultant axial force, based on the kinematic assumption
of inextensible PMA fibers. The principle of virtual work and total Lagrange
formulation are used to derive the element internal force vector and tangent
stiffness matrix. The PMA element is implemented in a geometrically nonlinear,
transient finite element program m for simulating the structural dynamics of
airdrop systems. Several numerical examples are given to validate the mechanical
behavior of the new element. Two large-scale application problems are also presented
to demonstrate the capabilities of the new element for simulating PMAs used
in airdrop systems.
In Zhou et al. (2004) has described, position
control problem of a two-degree-of-freedom arm system having a flexible second
link with artificial pneumatic muscle-type actuators. In Fig.
9, a composite controller design method is proposed in the framework of
the singular perturbation method. Various robust control schemes are designed
in order to meet with payload variation, parameter uncertainty, unmodeled vibration
mode, actuator dynamics both in the slow and the fast subsystems.
The PAMs has a wide choice for application in automation, robotics and material handling device. The construction and the capabilities of PAMs have been analyzed by various authors by adopting experimental and numerical methods. Apart from finding the characteristic of the PAMs, authors have indicated different strategies for analyzing the behaviors under different applications. Various techniques of modeling like Artificial Neural Networks (ARNN) and Fuzzy modeling system have been adopted for analyzing the rigidity and non-linear behavior of PAMs for getting good performance in controlling the PAM actuator for the real time industrial and human friendly applications.
The authors wish to thank the Universiti Teknologi PETRONAS (UTP) for the facilities to conduct the research.