UBIQUITOUS FRACTIONAL ORDER

CONTROLS?

YangQuan Chen ,1

Center for Self-Organizing and Intelligent Systems
(CSOIS), Department of Electrical and Computer
Engineering, Utah State University, Logan, UT
84322-4160, USA.

Abstract: There is an increasing interest in dynamic systems and controls of

noninteger orders or fractional orders. Clearly, for closed-loop control systems,
there are four situations. They are 1) IO (integer order) plant with IO controller;
2) IO plant with FO (fractional order) controller; 3) FO plant with IO controller
and 4) FO plant with FO controller. However, from engineering point of view,
doing something better is the major concern. This review article will first show
two examples that the best fractional order controller outperforms the best integer
order controller. Then, we try to argue why consider fractional order control even
when integer (high) order control works comparatively well. We will also address
issues in fractional order PID controller tuning. Using several real world examples,
we further argue that, fractional order control is ubiquitous when the dynamic
system is of distributed parameter nature.

Keywords: Non-integer order calculus, fractional order dynamic systems,

fractional order control, fractional order PID control, distributed parameter
systems, smart materials.

1. INTRODUCTION applying fractional calculus in dynamic systems

After Newton and Leibniz discovered calculus in
the 17th century, fractional-order calculus has
been studied as an alternative calculus in mathe-
matics (Debnath, 2004; Magin, 2004). As claimed
in (Chen et al., 2004c), fractional order calculus
will play an important role in mechatronic and
biological systems. Recently, in control society,
fractional order dynamic systems and controls
have gained an increasing attention (Lurie, 1994;
Podlubny, 1999; Oustaloup et al., 1995, 1996; Ray-
naud and Zerganoh, 2000). Pioneering works in
1 Corresponding author: Dr YangQuan
Chen. E-mail: yqchen@ece.usu.edu; Tel. 01-
435-7970148; Fax: 01-435-7973054. URL:
http://www.csois.usu.edu/people/yqchen.
and controls and the recent developments can be
found in (Manabe, 1960, 1961; Oustaloup, 1981;
Axtell and Bise, 1990; Vinagre and Chen, 2002;
Editor), 2002; Ortigueira and Editors), 2003). For
more detailed explanation about the fractional
dynamics and control, refer to (Xue and Chen,
2002).
Clearly, for closed-loop control systems, there are
four situations. They are 1) IO (integer order)
plant with IO controller; 2) IO plant with FO
(fractional order) controller; 3) FO plant with IO
controller and 4) FO plant with FO controller.
From engineering point of view, doing something
better is the major concern. This review arti-
cle will first show two examples that the best
fractional order controller outperforms the best
integer order controller. Then, we try to argue
why consider fractional order control even when
integer (high) order control works comparatively
well. We will also address issues in fractional order
PID controller tuning. Using several real world
examples, we further argue that, fractional order
control is ubiquitous when the dynamic system is
Fig. 1 shows the responses to unit step of the
angular position controlled by two integer order
PID controllers Gc1 (s) and Gc2 (s), respectively
with the Bode plots of the open-loop controlled
system shown in the same figure.
Step Response
100
Bode Diagram

of distributed parameter nature.

Magnitude (dB)
1.5 50

This review article is structured as follows. In 1 50

Amplitude
100
Sec. 2, we give the first example to show that the 0.5
360

Phase (deg)
180

best fractional order PID controller outperforms 0

the best integer order PID controller for an integer 0 180

360
2 1 0 1 2
0 20 40 60
order plant (DC motor with elastic shaft). Then, Time (sec)
10 10 10
Frequency (rad/sec)
10 10

in Sec. 3, the second example is presented to (a) Step responses (b) Bode plots
show that, for a fractional order plant, the best
fractional order boundary controller outperforms Fig. 1. Best IO PID Controllers. Solid line: ITAE;
the best integer order boundary controller. We Broken line: ISE
devoted a dedicated section, Sec. 4, on fractional
Now let us look at the best FO PID controllers.
order PID controllers and the related issues. Sec-
As the first attempt, let us first fix = 0.5
tion 5 introduces three other fractional order con-
and = 0.6. Doing the numerical search, we
trollers, namely, TID (tilted integral derivative)
get the best ITAE of 2.22 and the corresponding
controller, CRONE controller and fractional lead-
fractional order PID controller is
lag compensator. In Sec. 6, we focus on justify-
ing that fractional order dynamics and controls 0.01
Gc3 (s) = 135.12 + 31.6s0.6 (3)
are ubiquitous for distributed parameter systems s0.7
with an emphasis on smart material based sensing
and actuation for biomimetics. A conjecture is while the best ISE is 0.87 and the corresponding
presented in the end of Sec. 6 on the possible fractional order controller is
potential of fractional order controllers for inde- 91.95
pendent loopshaping for gain and phase. Finally, Gc (s) = 61.57 + + 2.33s0.6 . (4)
s0.5
Sec. 7 concludes this paper by offering some future
perspectives on fractional order controls. The step responses are compared in Fig. 2 with
corresponding Bode plots.
2. EXAMPLE-1: FO CONTROLLER FOR IO Step Response
100
Bode Diagram
Magnitude (dB)

PLANT 1.2
50
0
1
50
Amplitude

0.8
100
In this section, we focus on using FO-PID con- 0.6 150
360

troller for an IO plant - DC-Motor with elas- 0.4

Phase (deg)

180
0.2 0
tic shaft, a benchmark system from (Mathworks 0 180

Inc., 2006). Detailed results can be found in (Xue 0.2

0 10 20 30
Time (sec)
40 50 60
360
10
2
10
1
10
0

Frequency (rad/sec)
10
1
10
2

et al., 2006). (a) Step responses (b) Bode plots

2.1 Best IO PID vs. Best FO PID Fig. 2. Best FO PID Controllers. Solid line: ITAE;
Broken line: ISE
We used constrained optimization routine to The observation is clear. The best FO PID per-
search for the best controller parameters. Two forms better than the best IO PID. This is not
optimization criteria are used. One is ITAE (inte- surprising but this may not be fair since FO PID
gral of time-weighted absolute error) and another has two more extra parameters in optimal search.
one is ISE (integral of squared error), where the
constraint is the maximum torque less than Nm.
The reference signal is the unit step function.
2.2 How To Decide and ?
For the optimally searched IO PID using ITAE,
21.13 In the last section, we got a flavor that FO PID
Gc1 (s) = 41.94 + 8.26s; (1) performs better in side by side comparison. We
s
simply fixed = 0.5 and = 0.6. But in reality,
For optimally searched IO PID using ISE, how to decide these two magic orders? To our
10.65 best knowledge, this research question is not well
Gc2 (s) = 110.09 + + 30.97s (2) answered in the literature. Here, we only show a
s
brutal force search result to partially justify why ux (1, t) = f (t), (7)
we fixed = 0.5 and = 0.6.
u(x, 0) = u0 (x), (8)
Here, we build two tables of optimal ITAE and
ISE, respectively, with respect to and which ut (x, 0) = v0 (x), (9)
are enumerated from 0.5 to 1.5 with step of 0.1. In
other words, we did 2 11 11 optimal searches. where u(x, t) is the displacement of the cable at
These two tables are visualized in Fig. 3. x [0, 1] and t 0, f (t) is the boundary control
force at the free end of the cable, u0 (x) and
50 8 v0 (x) are the initial conditions of displacement
40

30
6 and velocity, respectively.
4
20

10 2 The control objective is to stabilize u(x, t), given

0
1.5
1.3
0
1.5
1.3
the initial conditions (8) and (9).
1.5 1.5
1.1 1.3 1.1 1.3
1.1 1.1
0.9
0.7
0.5 0.5
0.7
0.9
0.9
0.7
0.5 0.5
0.7
0.9
We adopt the following definition for the frac-
tional derivative of order of function f (t) (see
(a) ITAE(,) (b) ISE(,) (Mainardi and Paradisi, 1997) and (Mainardi,
1996)),
Fig. 3. Searching the best fractional orders (N = ( (n)
f (t) if = n N,
4) d .
f (t) = tn1 (10)
dt f (n) (t) if n 1 < < n,
(n )
Note that, in this investigation, we used the ap-
proximate order N = 4. It is unfortunate to ob-
where the denotes the time convolution between
serve that there is no definite relationship can be
two functions.
established in Fig. 3. However, in general, we can
qualitatively tell that, the integer case = 1 and In this paper, we study the performance and
= 1 is not optimal. In other words, the optimal properties of controllers in the following format:
case most likely corresponds to noninteger case.
Moreover, we can tell that, in this benchmark d u(1, t)
f (t) = k , 0<1 (11)
system, we prefer low order integral and lower dt
order derivative, which according to Fig. 3 give
better performance. where k is the controller gain, is the order of
fractional derivative of the displacement at the
free end of the cable.
3. EXAMPLE-2: FO CONTROL FOR FO When = 1, the controller (11) is called integer
PLANT order controller and has been widely used in the
boundary control of wave equations and beam
In this section, we consider a class of evolution equations (see (Chen, 1979), (Conrad and Morgul,
systems described by the one-dimensional time- 1998), and (Chen et al., 1987)). The effectiveness
fractional wave equation subject to a fractional or- has also been verified when applied to the bound-
der boundary controller. Via hybrid symbolic and ary control of fractional wave equation in (Liang
numerical simulation and parameter optimization, et al., 2004a). When 0 < < 1, can controller
we confirmed that the fractional order boundary (11) stabilize the system? What advantages does a
controller not only can stabilize the fractional fractional order controller have over integer order
wave equation, but performs better than an in- controllers?
teger order boundary controller as well. Detailed
results can be found in (Liang et al., 2004b, 2005).
3.2 Performance Comparison

3.1 Problem Formulation To test if fractional order boundary controllers can

be used to stabilize the fractional wave equation,
We consider a cable made with special smart ma- the following three different systems were simu-
terials governed by the fractional wave equation, lated
fixed at one end, and stabilized by a boundary
controller at the other end. Omitting the mass of Case 1: = 1.1, k = 0.1, = 0.5,
the cable, the system can be represented by Case 2: = 1.5, k = 0.1, = 0.7,
Case 3: = 1.9, k = 0.2, = 0.9.
u 2u
= , 1 < 2, x [0, 1], t 0 (5) All cases have the same initial conditions
t x2
u(0, t) = 0, (6) u0 (x) = sin(0.5x), v0 (x) = 0. (12)
Remark 3.1. In Case 3, k = 0.2 were chosen 0.4

rather than k = 0.1, because when the fractional =0.5

=0.7
0.2 =0.9
wave equation is closer to the wave equation (
1), the response tends to oscillate and needs a 0

larger damping to stabilize quicker.

0.2

u(1,t)
The descriptions of the simulation results are 0.4

summarized as follows:
0.6

For = 1.1, the displacement of the free end

is shown in Fig. 4. The displacement of the 0.8

whole cable for = 0.7 is shown in Fig. 5. 1

0 1 2 3 4 5 6 7 8 9 10
For = 1.5, the displacement of the free end t

is shown in Fig. 6. The displacement of the

whole cable for = 0.7 is shown in Fig. 7. Fig. 6. Displacement of the free end, = 1.5
For = 1.9, the displacement of the free end
is shown in Fig. 8. The displacement of the
whole cable for = 0.7 is shown in Fig. 9.
0.4

=0.5
=0.7
0.2 =0.9

0.2
u(1,t)

0.4

0.6

0.8
Fig. 7. Displacement of the whole cable, = 0.7
1
0 1 2 3 4 5 6 7 8 9 10 0.6
t

0.4 =0.5
=0.7
Fig. 4. Displacement of the free end, = 1.1 =0.9
0.2

0
u(1,t)

0.2

0.4

0.6

0.8

1
0 2 4 6 8 10 12 14 16 18 20
t

Fig. 8. Displacement of the free end, = 1.9

Fig. 5. Displacement of the whole cable, = 0.7

The simulation results show that all simulated

boundary controllers can stabilize the systems.
It is also shown that smaller means smaller
overshoot and longer rise time, and vice versa.
Next, we will study the performance comparison
between the fractional order boundary controller
and the integer order boundary controller.
First, we will find the optimal fractional boundary
controller applied to the classical wave equation
( = 1) using the initial condition (12). k and Fig. 9. Displacement of the whole cable, = 0.7
 are initialized as k = 1 and = 0.5. The 0.2

optimal values of k and turn out to be k =

1, = 1, which means the integer order bound- 0

ary controller achieves the best performance. The

displacement of the free end and the whole cable 0.2

are shown in Fig.10 and Fig.11, respectively.

u(1,t)
0.4

0.2

0.6
0 optimal fractional order controller
optimal integer order controller

0.8
0.2

0.4 1
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
u(1,t)

0.6

Fig. 12. Comparison between two optimal bound-

0.8
ary controllers for = 1.5
1

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

t

Fig. 10. Displacement of the free end, = 1, k =

1, = 1

Fig. 13. Displacement of the whole cable, = 1.5,

optimal fractional order boundary controller

Fig. 11. Displacement of the whole cable, =

1, k = 1, = 1

We can see that the response becomes zero for

t > 2, an already well-known result (see (Chen
and Zhou, 1990)).
Is the integer order boundary controller always
better than the fractional order boundary con-
troller? Let us try the fractional wave equation
with = 1.5. The optimal fractional order con-
troller turns out to be k = 0.7608 and =
0.9275. The optimal integer order boundary con-
troller is with gain k = 1.1453.
The comparison of the free end displacement
between the optimal fractional order boundary
controller and optimal integer order boundary
controller is shown in Fig. 12. From Fig. 12 we
can see that the response to the optimal fractional
order boundary controller not only has shorter rise
time and settling time, but also has no overshoot.
The plots of the displacement of the whole cable
are shown in Fig. 13 and Fig. 14, respectively.
Fig. 14. Displacement of the whole cable, = 1.5,
optimal integer order boundary controller
Finally, we study the case of = 1.1, in which
the fractional wave is much closer to the diffusion
equation than to the wave equation. The optimal
fractional order controller is k = 0.2455, =
0.8882 and optimal integer order controller is
k = 0.6787. The comparison of the free end
displacement is shown in Fig. 15. We can see that
the optimal fractional order boundary controller is
again much better than the optimal integer order
boundary controller. The displacements of the
whole cable are comapred in Fig. 16 and Fig. 17,
respectively.
 0.2 Subject to: k > 0 and 0 < 1. In the above
optimization tasks, u(1, t) is the displacement of
0
the free end of the cable; tf is the total time of
simulation; T is the time period to optimize within
0.2
the time interval [tf T, tf ] which is determined
by trial-and-error.
u(1,t)

0.4

optimal fractional order controller

optimal integer order controller
The optimization program we chose is SolvOpt
0.6 (see (Kuntsevich and Kappel, 1997)), a free pro-
gram for local nonlinear optimization problems.
0.8

1
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 4. FRACTIONAL ORDER PID
t
CONTROLLERS
Fig. 15. Comparison between two optimal bound-
ary controllers for = 1.1 According to a survey (Yamamoto and Hashimoto,
1991) of the state of process control systems in
1989 conducted by the Japan Electric Measur-
ing Instrument Manufacturers Association, more
than 90 percent of the control loops were of the
PID type. It was also indicated (Bialkowski, 1994)
that a typical paper mill in Canada has more than
2,000 control loops and that 97 percent use PI con-
trol. Therefore, the industrialist had concentrated
on PI/PID controllers and had already developed
one-button type relay auto-tuning techniques for
fast, reliable PI/PID control yet with satisfactory
performance (Leva, 1993; Karl J. Astrom, 1995;
Yu, 1999; Tan et al., 2000; Aniruddha Datta,
2000).
Fig. 16. Displacement of the whole cable, = 1.1, Intuitively, with noninteger order controllers for
optimal fractional order boundary controller integer order plants, there is a better flexibility in
adjusting the gain and phase characteristics than
using IO controllers. This flexibility makes FO
control a powerful tool in designing robust control
system with less controller parameters to tune.
The key point is that using few tuning knobs, FO
controller achieves similar robustness achievable
by using very high-order IO controllers.
PI D controller, also known as PI D controller,
was studied in time domain in (Podlubny, 1999)
and in frequency domain in (Petras, 1999). In
general form, the transfer function of PI D is
given by
U (s)
Fig. 17. Displacement of the whole cable, = 1.1, C(s) = = Kp + Ti s + Td s , (15)
E(s)
optimal integer order boundary controller

For integer order boundary controllers ( = 1), where and are positive real numbers; Kp is the
we seek the best gain k to proportional gain, Ti the integration constant and
Td the differentiation constant. Clearly, taking
min J(k) = max abs((u(1, t))), t [tf T, tf ] (13) = 1 and = 1, we obtain a classical PID
k
controller. If = 0 (Ti = 0) we obtain a PD
controller, etc. All these types of controllers are
Subject to: k > 0. particular cases of the PI D controller. The time
domain formula is that
For fractional order boundary controllers (0 <
() ()
1), the task to to find the best gain k and u(t) = Kp e(t) + Ti D
t e(t) + Td Dt e(t). (Dt 0 Dt ).(16)
the fractional order to
It can be expected that PI D controller (16) may
min J(k, ) = max(u(1, t)), t [tf T, tf ](14) enhance the systems control performance due to
k,
more tuning knobs introduced. Actually, in the-
ory, PI D itself is an infinite dimensional linear
filter due to the fractional order in differentiator or
integrator. For controller tuning techniques, refer
to (Monje et al., 2004; Chen et al., 2004a).
Similar to the fact that, every year, numerous
PI/PID papers have been published, we can fore-
see that, more and more FO PI/D papers will be
published in the future. In general, the following
issues should be addressed:
How to tell there is a need to use FO PI/D
controller while integer order PI/D control
works well in the existing controlled systems.
How to predict the performance gain by using
FO PI/D controller?
How to best tune the FO PI/D controller by
taking minimum experimental efforts?
How to best design the experiments to tune
FO PI/D controller?
For a given class of plants to be controlled,
how to best design FO PI/D controller?
In the interest of space, we conclude this section
by referring to the two recent Ph.D. dissertations
(Zhao, 2006; Concepcion Alicia Monje Micharet,
2006) and the references therein and a recent talk
slides (Concepcion Alicia Monje Micharet, 2005)
for more recent results on fractional order PID
controllers.
Remark 4.1. We comment that since PID control
is ubiquitous in industry process control, FO PID
control will be also ubiquitous when tuning and
implementation techniques are well developed.
5. SOME OTHER FRACTIONAL ORDER
CONTROLLERS
As already widely known, the early attempts
to apply fractional-order derivative to systems
control can be found in (Manabe, 1960; Axtell
and Bise, 1990; Oustaloup, 1981). In this section,
three other representative fractional-order con-
trollers in the literature will be briefly introduced,
namely, TID (tilted integral derivative) controller,
CRONE controller and fractional lead-lag com-
pensator. For detailed introduction and compari-
son, refer to (Xue and Chen, 2002). For the latest
developments, we refer to (Vinagre and Chen,
2002; Mehaut et al., 2004; Oustaloup, 2006).
TID Controller. In (Lurie, 1994), a feedback
control system compensator of the PID type
is provided, wherein the proportional com-
ponent of the compensator is replaced with
a tilted component having a transfer func-
1
tion s n . The resulting transfer function of
the entire compensator more closely approx-
imates an optimal transfer function, thereby
achieving improved feedback controller. Fur-
ther, as compared to conventional PID com-
pensators, the TID compensator allows for
simpler tuning, better disturbance rejection
ratio, and smaller effects of plant parameter
variations on closed loop response.
The objective of TID is to provide an
improved feedback loop compensator having
the advantages of the conventional PID com-
pensator, but providing a response which is
closer to the theoretically optimal response.
In TID patent (Lurie, 1994), an analog circuit
using op-amps plus capacitors and resistors
is introduced with a detailed component list
which is useful in some cases where the com-
puting power to implementing T3 (s) digitally
is not possible. An example is given in (Lurie,
1994) to illustrate the benefits from TID over
conventional PID in both time and frequency
domain.
CRONE Controller. The CRONE con-
trol was proposed by Oustaloup in pur-
suing fractal robustness (Oustaloup et al.,
1995, 1996). CRONE is a French abbrevia-
tion for Controle Robuste dOrdre Non En-
tier (which means non-integer order robust
control). In this section, we shall follow the
basic concept of fractal robustness, which mo-
tivated the CRONE control, and then mainly
focus on the second generation CRONE con-
trol scheme and its synthesis based on the
desired frequency template which leads to
fractional transmittance (Oustaloup et al.,
1999; Oustaloup and Mathieu, 1999).
In (Oustaloup, 1991), fractal robustness
is used to describe the following two char-
acteristics: the isodamping and the vertical
sliding form of frequency template in the
Nichols chart. This desired robustness mo-
tivated the use of fractional-order controller
in classical control systems to enhance their
performance.
With a unit negative feedback, for the
characteristic equation
1 + ( s) = 0,
the forward path transfer function, or the
open-loop transmittance, is that
   
1 u
(s) = = , (17)
s s
which is the transmittance of a non integer
integrator in which u = 1/ denotes the
unit gain (or transitional) frequency.
In controller design, the objective is to
achieve such a similar frequency behavior, in
a medium frequency range around u , know-
ing that the closed loop dynamic behavior is
exclusively linked to the open loop behavior
around u . Synthesizing such a template de-
 fines the non-integer approach that the sec-
ond generation CRONE control uses.
There are a number of real life applications
of CRONE controller such as the car sus-
pension control (Oustaloup, 1990; Oustaloup
et al., 1996), flexible transmission (Oustaloup
et al., 1995), hydraulic actuator (Lanusse
et al., 2000) etc. CRONE control has been
evolved to a powerful non-conventional con-
trol design tool with a dedicate MATLAB
toolbox for it (Oustaloup et al., 2000b). For
an extensive overview, refer to (Oustaloup
et al., 2000a) and the references therein.
Fractional Lead-Lag Compensator. In the
above, fractional controllers are directly re-
lated to the use of fractional-order differ-
entiator or integrator. It is possible to ex-
tend the classical lead-lag compensator to
the fractional-order case which was studied in
(Raynaud and Zerganoh, 2000; IFA, 2005).
The fractional lead-lag compensator is given
by
 r
1 + s/b
Cr (s) = C0 (18)
1 + s/h
where 0 < b < h , C0 > 0 and r (0, 1).
The autotuning technique has been presented
in (IFA, 2005).
We conclude this section by offering the following
remark.
Remark 5.1. Just like the non-integers are ubiq-
uitous between integers, noninteger order con-
trol schemes will be ubiquitous by extending the
existing integer order control schemes into their
noninteger counterparts. For example, fractional
sliding mode control with fractional order slid-
ing surface dynamics; model reference adaptive
control using fractional parameter updating law
etc. The opportunities for extensions are almost
endless. However, the question remains: we need
a good reasons for such extensions. Performance
enhancement as demonstrated in previous sections
is only part of the reason. There are some deeper
reasons as discussed in the next section.
6. UBIQUITOUS FRACTIONAL ORDER
CONTROLS
Let us begin by referring to (Desai, 2000). In the
preface of (Desai, 2000), Desai remarked Stu-
dents of mechanics of materials often raise the
question, Is there a constitutive model which is
applicable to all materials? And I respond Al-
though our understanding of the materials re-
sponse is growing, there is no model available that
can characterize all materials in all respects. To
understand and characterize matter (materials)
completely, one may need to become the mat-
ter itself! When that happens, there is no differ-
ence left, and a full understanding may follow.
This realization is important because the pursuit
towards increased comprehension and improved
characterization of materials must continue! In
other words, our world is distributed parameter
in nature. As our technology evolves, the lumped
parameter thinking about our world is no longer
efficient enough for increasingly demanding per-
formance.
For distributed parameter systems (DPS), it has
been shown that fractional order calculus will play
a role in its modeling and analysis. It is natural
to consider fractional order controls. In between
DPS and lumped parameter systems (LPS), there
are other types of systems such as porous media
based systems and particulate or granular sys-
tems. Depending on the scales of modeling (micro,
meso, macro), fractional order dynamics will show
up at certain scale. We can conclude that, prob-
ably, fractional calculus might be the right tool
to characterize the DPS using the LPS viewpoint.
In short, DPS is ubiquitous, so is fractional
order calculus.
On the other hand, smart materials have been
developed to such a state that biomimetics is
possible using smart material based sensing and
actuation. Just imagine our central nerve sys-
tem and the neuromusculoskeletal system. It is
distributed parameter in nature and has infinite
dimensional dynamics with memory. It is analog
in nature, too. Using digital modeling is just for
our own convenience. Likewise, using integer or-
der dynamic modeling is only for our own conve-
nience. We believe, fractional order calculus may
be the right notion to modeling and control of
neuromusculoskeletal systems.
As we know that biomimetics is an engineer-
ing discipline learning from nature. Biomimetics
is in fact an enabling discipline which looks to-
wards nature for ideas that may be adapted and
adopted for solving problems: Inspiration rather
than imitation. Although in the current stage,
the dominant efforts are on characterizing the
biomimetic material properties, eventually, the
control will come into play. The typical behav-
iors of biomimetic actuators include the time-
dependent relaxation, anisotropy, elasticity, vis-
coelasticity, creep and hysteresis and other memo-
ryless nonlinearities. These are tough challenging
behaviors for advanced controllers. To appreci-
ate the importance of controls, let us refer to
the recent special issue on dynamics and con-
trol of smart structures, IEEE Transactions on
Control Systems Technology. The Guest Edito-
rial (Reza Moheimani and Goodwin, 2001) tells
that Materials that are used in the construction
of smart structures include, but are not limited
to, piezoelectric materials, shape memory alloys,
magnetorheological and electrorheological fluids
and magnetostrictive materials. Most of these
materials have been known for a long time. For
example, piezoelectricity was first discovered in
the early 1800s. The reason for an explosion of
recent interest in this area can be attributed to
the availability of powerful computers over the
past two decades, which has allowed researchers
and engineers to control the behavior of smart
structural systems in efficient ways. A key fac-
tor in guaranteeing high-performance operation of
smart structures is the control algorithm. In this
special issue, many papers tried to attack the hys-
teresis problem (Cruz-Hernandez and Hayward,
2001; Majima et al., 2001, e.g.) since hysteresis
is an intrinsic characteristic of transducers based
on smart materials, which rely on modifications
of stress-strain relationships invariably associated
with hysteretic behavior.
We briefly envision the following on fractional
calculus for biomimetic control:
Fractional order modeling and fractional or-
der system identification. To better model
the dynamics of biomimetic materials, inte-
ger order based model will not be enough.
Furthermore, the fractional order itself could
be varying or state-dependent. Refer to
(Hartley and Lorenzo, 2003) and the refer-
ences therein for more details.
Fractional order controller design. In pre-
vious sections, it has been shown that frac-
tional order controllers can improve the con-
trol systems performance. However, the con-
troller design and tuning methods are just of
a recent focus (Monje et al., 2004; Chen et al.,
2004a). The fractional order disturbance ob-
server (Chen et al., 2003) will be useful for
vibration suppression in biomimetic control
systems.
Fractional order controller for nonlinearity
compensation. For memoryless nonlineari-
ties such as deadzone, backlash, static distor-
tion can be compensated by using fractional
order control. For example, (Ma and Hori,
2004) investigated the use of FOC to sup-
press the vibration due to backlash in mo-
tion control systems. For nonlinearities with
memory, the typical one is the hysteresis. It
is for sure that FOC can compensate the
hysteresis effect in a closed-loop control sys-
tems. This is motivated by observing the Fig.
10 of (Cruz-Hernandez and Hayward, 2001)
where the so-called phaser is nothing but
a band-pass approximation of a fractional
order differentiator.
Having pointed out the possible use of FOC in
biomimetic control, it is now important to have
a look of the implementation issue of fractional
order controllers. Existing methods are all based
on certain approximation schemes in either digital
(Oustaloup et al., 2000b; Chen et al., 2004b) form
or analog form (Petras et al., 2002). Another im-
portant direction for FOC realization is the au-
thentic application using Fractors (Bohannan,
2002). The key idea of Fractor is that we can
make use of certain lossy dielectrics to make a
capacitance with impedance of the form of ZF =
K/s with K the lumped gain and [0, 1] the
fractional order (Bohannan, 2002). For example,
the lithium hydrazinium sulphate (LiN2 H5 SO4 )
has such an impedance characteristic. This idea is
now currently under pursuing aiming to mass pro-
duce compact FOC elements. We can expect that
a revolution may happen when existing PI/PID
controllers could be upgraded to fractional order
PI/PID controllers since more than 90% of the
industrial controllers are of PI/PID type.
Since the dynamic behaviors of many biomimetic
materials exhibit certain fractionality, this prompts
us to check the other suitable implementations
of FOC using biomimetic materials similar to
LiN2 H5 SO4 . Therefore, the big picture for the fu-
ture is the intelligent control of biomimetic system
using biomimetic materials with fractional order
calculus embedded. In other words, it is definitely
worth to have a look of the notion of intelligent
control of intelligent materials using intelligent
materials. Advocating this picture is one of the
major purposes or contributions, if any, of this
paper.
Finally, we conclude this section by presenting the
following conjecture:
Remark 6.1. Conjecture: It is well known due
to Bode that, for finite dimensional linear time
invariant single input and single output rational
systems, the gain and phase are inter-related. For
robust loop shaping, it is not possible to do loop-
shaping for gain and loop phase plots indepen-
dently. However, using fractional order control,
this is possible. In other words, fractional order
controllers have the potential to do loopshaping
of phase and gain independently.
7. CONCLUSIONS AND FUTURE
PERSPECTIVE
In this review article, we first showed two exam-
ples that the best fractional order controller out-
performs the best integer order controller. Then,
we tried to argue why consider fractional order
control even when integer (high) order control
works comparatively well. We also addressed is-
sues in fractional order PID controller tuning.
Using several real world examples, we further ar-
gue that, fractional order control is ubiquitous
when the dynamic system is of distributed pa-
rameter nature. In particular, we discussed the
biomimetic control and the role of fractional order
control in biomimetics. We highlighted the notion
of intelligent control of intelligent materials using
intelligent materials. We also conjectured that
fractional order controllers have the potential to
do loopshaping of phase and gain independently.
As for the future perspective, we briefly offer the
following remarks for future investigation
Power law Lyapunov stability theory should
replace the exponential law based Lyapunov
stability theory?
Power law phenomena are due to time-
/spatial-fractional order dynamics?
Time frequency analysis, multi-resolution
analysis (wavelet), fractional Fourier trans-
formation, and fractional order calculus are
inter-related?
Long range dependence of stochastic process
is due to fractional order dynamics?
In IEEE CDC2002 Tutorial Workshop,
Some of our investigations, to be published else-
where, have already shown that, some of the above
speculations are true.
As a final remarks, the readers are reminded that
whenever the following words appears:
power law
long range dependence
porous media
particulate
granular
lossy
anomaly
disorder
YangQuan Chen, B. M. Vinagre, and Igor Pod-
think about fractional order dynamics and con-
trols. In general, fractional order dynamics and
controls are ubiquitous.
ACKNOWLEDGMENTS
The author would like to thank his coauthors
Professor Igor Podlubny, Professor Dingyu Xue,
Professor Blas Vinagre, Dr. Jinsong Liang, Dr.
Hyosung Ahn, Dr. Chunna Zhao and Dr. Concha
Monje. The research reviewed in this paper was
supported in part by Utah State University New
Faculty Research Grant, the TCO Bridging Fund
of Utah State University (2005-2006), an NSF
SBIR subcontract through Dr. Gary Bohannan,
and by the National Academy of Sciences under
the Collaboration in Basic Science and Engineer-
ing Program/Twinning Program supported by
Contract No. INT-0002341 from the National Sci-
ence Foundation. The contents of this publication
do not necessarily reflect the views or policies of
the National Academy of Sciences or the National
Science Foundation, nor does mention of trade
names, commercial products or organizations im-
ply endorsement by the National Academy of Sci-
ences or the National Science Foundation.
REFERENCES
Shankar P. Bhattacharyya Aniruddha Datta,
Ming-Tzu Ho. Structure and Synthesis of PID
Controllers. Springer-Verlag, London, 2000.
M. Axtell and E. M. Bise. Fractional calculus ap-
plications in control systems. In Proceedings of
the IEEE 1990 Nat. Aerospace and Electronics
Conf., pages 563566, New York, USA, 1990.
W. L. Bialkowski. Dreams versus reality: A view
from both sides of the gap. Pulp and Paper
Canada, 11:1927, 1994.
Gary W. Bohannan. Analog realization of
a fractional control element - revisited.
http://mechatronics.ece.usu.edu/foc/cdc02tw/
Las Vegas, NE, USA, 2002.
G. Chen. Energy decay estimates and exact
boundary value controllability for the wave
equation in a bounded domain. J. Math. Pure.
Appl., 58:249273, 1979.
G. Chen and J. Zhou. The wave propagation
method for the analysis of boundary stabiliza-
tion in vibrating structures. SIAM J. Appl.
Math., 50(5):12541283, 1990.
G. Chen, M. C. Delfour, A. M. Krall, and
G. Payre. Modelling, stabilization and control
of serially connected beams. SIAM J. Contr.
Optimiz., 25:526546, 1987.
lubny. On fractional order disturbance ob-
servers. In Proc. of The First Symposium on
Fractional Derivatives and Their Applications
at The 19th Biennial Conference on Mechan-
ical Vibration and Noise, the ASME Interna-
tional Design Engineering Technical Confer-
ences & Computers and Information in Engi-
neering Conference (ASME DETC2003), pages
18, DETC2003/VIB--48371, Chicago, Illinois,
2003.
YangQuan Chen, Kevin L. Moore, Blas M. Vina-
gre, and Igor Podlubny. Robust PID controller
autotuning with a phase shaper. In Proceedings
of The First IFAC Symposium on Fractional
Differentiation and its Applications (FDA04),
Bordeaux, France, 2004a.
YangQuan Chen, Blas M. Vinagre, and Igor
Podlubny. Continued fraction expansion ap-
proaches to discretizing fractional order deriva-
 tives - an expository review. Nonlinear Dynam-
ics (Kluwer), 9(1):to appear, 2004b.
YangQuan Chen, Dingyu Xue, and Huifang Dou.
Fractional calculus and biomimetic control. In
Proc. of the First IEEE Int. Conf. on Robotics
and Biomimetics (RoBio04), pages robio2004
347, Shengyang, China, August 2004c. IEEE.
Concepcion Alicia Monje Micharet. Auto-tuning
of fractional PID controllers. In IEEE Con-
trol System Society San Diego Chapter Meet-
ing. In Trex Enterprises,, page Slides available
at http://mechatronics.ece.usu.edu/foc/,
San Diego, California, USA, Sept. 29 2005.
Concepcion Alicia Monje Micharet. Design Meth-
ods of Fractional Order Controllers for Indus-
trial Applications. PhD thesis, University of
Extremadura, Spain, 2006.
Francis Conrad and Omer Morgul. On the stabil-
ity of a flexible beam with a tip mass. SIAM
Journal of Control and Optimization, 36(6):
19621986, 1998.
J. M. Cruz-Hernandez and V. Hayward. Phase
control approach to hysteresis reduction. IEEE
Transactions on Control Systems Technology, 9
(1):1726, 2001.
Lokenath Debnath. A brief historical introduction
to fractional calculus. Int. J. Math. Educ. Sci.
Technol., 35(4):487501, 2004.
Chandrakant S. Desai. Mechanics of Materials
and Interfaces: The Disturbed State Concept.
CRC Press, ISBN 0-8493-0248-X TA405 .D45,
2000.
J. A. Tenreiro Machado (Guest Editor). Special
issue on fractional calculus and applications.
Nonlinear Dynamics, 29:1385, March 2002.
Tom T. Hartley and Carl F. Lorenzo. Fractional-
order system identification based on continuous
order-distributions. Signal Processing, 83(11):
2287 2300, 2003.
C. A. Monje, B. M. Vinagre, A. J. Calderon,
V. Feliu and Y. Q. Chen. Auto-Tuning of Frac-
tional Lead-Lag Compensators, Prague, July 4-8
2005. IFAC.
Tore Hagglund Karl J. Astrom. PID Controllers:
Theory, Design, and Tuning. ISA - The Instru-
mentation, Systems, and Automation Society
(2nd edition), 1995.
Alexei Kuntsevich and Franz Kappel. SolvOpt:
the solver for local nonlinear optimization prob-
lems, 1997.
P. Lanusse, V. Pommier, and A. Oustaloup. Frac-
tional control system design for a hydraulic
actuator. In Proc. of the First IFAC conference
on Mechatronics Systems, Mechatronic 2000,
Darmstadt, Germany, September 2000.
A. Leva. PID autotuning algorithm based on relay
feedback. IEEE Proc. Part-D, 140(5):328338,
1993.
J. Liang, Y. Q. Chen, and R. Fullmer. Simulation
studies on the boundary stabilization and dis-
turbance rejection for fractional diffusion-wave
equation. In Proceedings of the 2004 IEEE
American Control Conference, 2004a.
Jinsong Liang, YangQuan Chen, Blas M. Vinagre,
and Igor Podlubny. Boundary stabilization of a
fractional wave equation via a fractional order
boundary controller. In Proceedings of The
First IFAC Symposium on Fractional Differen-
tiation and its Applications, Bordeaux, France,
July 19-20, 2004b.
Jinsong Liang, YangQuan Chen, Blas M. Vinagre,
and Igor Podlubny. Fractional order boundary
stabilization of a time-fractional wave equation.
a chapter in Fractional Derivatives an Their
Applications. Editors: A. Le Mehaute, J. A.
Tenreiro Machado, J. C. Trigeassou and J.
Sabatier. UBooks , Augsburg, Germany, 2005.
Boris J. Lurie. Three-parameter tunable tilt-
integral-derivative (TID) controller. US Patent
US5371670, 1994.
Chengbin Ma and Yoichi Hori. The applications
of fractional order control to backlash vibration
supperssion. In Proceedings of The 2004 Amer-
ican Control Conference, Boston, USA, 2004.
Richard L. Magin. Fractional calculus in bio-
engineering. Critical ReviewsTM in Biomedical
Engineering, 32(1-4), 2004.
F. Mainardi. Fractional relaxation-oscillation and
fractional diffusion-wave phenomena. Chaos,
Solitons, and Fractals, 7(9):14611477, 1996.
F. Mainardi and P. Paradisi. A model of diffu-
sive waves in viscoelasticity based on fractional
calculus. In Proceedings of the 36th IEEE Con-
ference on Decision and Control, Hyatt Regency
San Diego, California, 1997.
S. Majima, K. Kodama, and T. Hasegawa. Mod-
eling of shape memory alloy actuator and track-
ing control system with the model. IEEE Trans-
actions on Control Systems Technology, 9(1):
5459, 2001.
S. Manabe. The non-integer integral and its
application to control systems. JIEE (Japanese
Institute of Electrical Engineers) Journal, 80
(860):589597, 1960.
S. Manabe. The non-integer integral and its
application to control systems. ETJ of Japan,
6(3-4):8387, 1961.
Mathworks Inc. Model Predictive
Control Toolbox Users Guide.
http://www.mathworks.com/access/helpdesk/
help/toolbox/mpc/, 2006.
A. Le Mehaut, J. A. Tenreiro Machado, J. C.
Trigeassou, and J. Sabatier, editors. Proceedings
of The First IFAC Symposium on Fractional
Differentiation and its Applications (FDA04),
Bordeaux, France, July 19-21 2004. IFAC, El-
sevier Science Ltd., Oxford, UK.
C. A. Monje, B. M. Vinagre, Y. Q. Chen, V. Fe-
liu, P. Lanusse, and J. Sabatier. Proposals
for fractional PI D tuning. In Proceedings
 of The First IFAC Symposium on Fractional
Differentiation and Its Applications (FDA04),
Bordeaux, France, 2004.
Manuel Duarte Ortigueira and J. A. Tenreiro
Machado (Guest Editors). Special issue on
fractional signal processing and applications.
Signal Processing, 83(11):22852480, Nov. 2003.
A. Oustaloup. Fractional order sinusoidal
oscilators: Optimization and their
use in highly linear FM modulators.
IEEE Transactions on Circuits and Systems,
28(10):10071009, 1981.
A. Oustaloup. La commande CRONE. Edition
Hermes, Paris, 1991.
A. Oustaloup. Nouveau systeme de suspension:
La suspension CRONE. INPI Patent 90 046
13, 1990.
A. Oustaloup, editor. Proceedings of The Second
IFAC Symposium on Fractional Differentiation
and its Applications (FDA06), Porto, Portugal,
July 19-21 2006. IFAC, Elsevier Science Ltd.,
Oxford, UK.
A. Oustaloup and B. Mathieu. La commande
CRONE : du scalaire au multivariable. HER-
MES, Paris, 1999.
A. Oustaloup, B. Mathieu, and P. Lanusse. The
CRONE control of resonant plants: application
to a flexible transmission. European Journal of
Control, 1(2), 1995.
A. Oustaloup, X. Moreau, and M. Nouillant. The
CRONE suspension. Control Engineering Prac-
tice, 4(8):11011108, 1996.
A. Oustaloup, J. Sabatier, and P. Lanusse. From
fractal robustness to CRONE control. Fraction-
nal Calculus and Applied Analysis, 2(1):130,
1999.
A. Oustaloup, F. Levron, F. Nanot, and B. Math-
ieu. Frequency band complex non integer differ-
entiator: characterization and synthesis. IEEE
Transactions on Circuits and Systems I: Funda-
mental Theory and Applications, 47(1):2540,
January 2000a.
A. Oustaloup, P. Melchoir, P. Lanusse, C. Cois,
and F. Dancla. The CRONE toolbox for Mat-
lab. In Proceedings of the 11th IEEE Inter-
national Symposium on Computer Aided Con-
trol System Design - CACSD, Anchorage, USA,
September 2000b.
I. Petras, I. Podlubny, P. OLeary, L. Dorcak, and
Vinagre B. Analogue Realization of Fractional
Order Controllers. FBERG, Technical Univer-
sity of Kosice, Kosice, Slovak, isbn 8070996277
edition, 2002.
I. Petras. The fractional-order controllers: meth-
ods for their synthesis and application. Jour-
nal of Electrical Engineering, 50(9-10):284288,
1999.
Igor Podlubny. Fractional-order systems and
PI D -controllers. IEEE Trans. Automatic
Control, 44(1):208214, 1999.
H.F. Raynaud and A. Zerganoh. State-space
representation for fractional order controllers.
Automatica, 36:10171021, 2000.
S. O. Reza Moheimani and G. C Goodwin. Guest
editorial introduction to the special issue on
dynamics and control of smart structures. IEEE
Transactions on Control Systems Technology, 9
(1):34, 2001.
Kok Kiong Tan, Wang Qing-Guo, Hang Chang
Chieh, and Tore Hagglund. Advances in PID
Controllers. Advances in Industrial Control.
Springer-Verlag, London, 2000.
Blas M. Vinagre and YangQuan Chen. Lecture
notes on fractional calculus applications
in automatic control and robotics. In
Blas M. Vinagre and YangQuan Chen,
editors, The 41st IEEE CDC2002 Tutorial
Workshop # 2, pages 1310. [Online]
http://mechatronics.ece.usu.edu
/foc/cdc02 tw2 ln.pdf, Las Vegas, Nevada,
USA, 2002.
Dingyu Xue and YangQuan Chen. A compara-
tive introduction of four fractional order con-
trollers. In Proc. of The 4th IEEE World
Congress on Intelligent Control and Automa-
tion (WCICA02), pages 32283235, Shanghai,
China, June 2002. IEEE.
Dingyu Xue, Chunna Zhao, and YangQuan Chen.
Fractional order PID control of a DC-motor
with elastic shaft: A case study. In Proc. of
the American Control Conference, Minneapolis,
MN, USA, June 2006.
S. Yamamoto and I. Hashimoto. Recent status
and future needs: The view from Japanese in-
dustry. In Arkun and Ray, editors, Proceed-
ings of the fourth International Conference on
Chemical Process Control, Texas, 1991. Chemi-
cal Process Control CPCIV.
Cheng-Ching Yu. Autotuning of PID Controllers:
Relay Feedback Approach. Advances in Indus-
trial Control. Springer-Verlag, London, 1999.
Chunna Zhao. Research on Analyse and Design
Methods of Fractional Order System. PhD
thesis, Northeastern University, China, 2006.