World Applied Sciences Journal 8 (5): 555-564, 2010

ISSN 1818-4952
IDOSI Publications, 2010

Corresponding Author: Dr. Mehdi Roopaei, Department of Electrical Engineering, Fars Research and Science-Islamic Azad
University, Shiraz, Iran
555
Car Suspension Control By Indirect Adaptive Interval Type-2
Fuzzy Neural Network Control

1
Tsung-Chih Lin,
2
Mehdi Roopaei and
1
Ming-Che Chen

1
Department of Electronic Engineering, Feng-Chia University, Taichung, Taiwan
2
Department of Electrical Engineering, Fars Research and Science-Islamic Azad University, Shiraz, Iran

Abstract: To control quarter-car systems, a novel indirect adaptive interval type-2 Fuzzy Neural Network
(FNN) controller is developed in this paper to achieve both ride comfort and good handling. By
incorporating indirect adaptive interval type-2 FNN control approach and sliding mode control, car
suspension system regulation performance can be achieved based on Lyapunov stability criterion. The
simulation example is given to confirm validity of the proposed design scheme.

Key words: Indirect adaptive control interval type-2 fuzzy FNN car suspension system

INTRODUCTION

The main role of a car suspension system is to
improve the ride comfort and to better the handling
property. A traditional spring-damper suspension is
currently being replaced by either a semi -active or
active suspension system. In order to improve the
capability of active suspension systems, many research
studies have been investigated on the design and control
of active suspension system recently [1-4].
FNN controller is defined as an FNN logic system
equipped with an adaptation algorithm. Moreover,
FNN is constructed from a collect of fuzzy IF-THEN
rules using fuzzy logic principles and the adaptation
algorithm adjusts the free parameters of the FNN based
on the numerical experiment data. Like the
conventional adaptive control, the adaptive FNN
control has direct and indirect FNN adaptive control
categories. Direct adaptive FNN control has been
discussed which the adaptive FNN controller uses fuzzy
logic systems as controller. Hence linguistic fuzzy
control rules can be directly incorporated into the
controller. Also, indirect adaptive FNN control has
been proposed which the indirect FNN controller uses
fuzzy descriptions to model the plant. Hence, fuzzy
IF-THEN rules describing the plant can be directly
incorporated into the indirect FNN controller [5-11].
In general, the rule uncertainty will be existed in
following three possible ways, (i) the words that are
used in antecedents and consequents of rules can mean
different things to different people; (ii) consequents
obtained by polling a group of experts will often be
different for the same rule because the experts will not
necessary be in agreement and (iii) noisy training
data. Therefore, antecedent or consequent uncertainties
translate into uncertain antecedent or consequent
membership functions. Type-1 FLSs are unable to
handle rule uncertainties directly, when the information
that is used to construct the rules in a FLS is uncertain.
On the other hand, type-2 FLSs involved in this paper
whose antecedent or consequent membership functions
are type-2 fuzzy sets can handle rule uncertainties. A
type-2 FLS is characterized by IF-THEN rules, but its
antecedent or consequent sets are type-2. Hence, type-2
FLSs can be used when the circumstances are too
uncertain to determine exact membership grades such
as when training data are corrupted by noise [12, 13].
Since, appropriate improvements of active
suspension systems have a potential to improve the ride
comfort and vehicle maneuverability, this research area
has remained attractive for many years. Chen et al. [14]
introduced a Fuzzy Logic Controller (FLC) for an
active suspension under the Sliding Mode Control
(SMC) approach. The FLC was assigned to reduce the
chattering of the SMC and to supervise the control
performance to self-tune parameters of the control
system. Yun et al. [15] also proposed a fuzzy adaptive
sliding mode controller for an active suspension system
with different sliding surface proposed by Chen et al.
Rao and Prahlad [16] also presented a adjustable FLC
to reduce the suspension acceleration and deflection to
a level of a hypothetical reference model. The
suspension deflection and velocity were the two input
variables to the FLC while the desired force was its
output. Fernando and Viassolo [17] presented a
controller consisting of two control loops to minimize
World Appl. Sci. J., 8 (5): 555-564, 2010
556

the vehicle bodys acceleration and to avoid hitting the
suspension limit. The outer loop was the same as that
proposed by Rao and Prahlad but an additional inner
loop was designed to deal with the non-linear hydraulic
actuator. More recent research results can be found in
references [18-21].
This paper is structured as follows. Car suspension
system is given in section II. A brief description of
interval type 2 fuzzy neural network is introduced in
section III.
Indirect adaptive interval type-2 fuzzy neural
network control for chaos synchronization between two
different chaotic systems is presented in section III.
Simulation example to demonstrate the performances of
the proposed method is provided in section IV. Section
V gives the conclusions of the advocated design
methodology.

QUARTER-CAR MODEL

In this paper, we are considering a quarter car
model with two degrees of freedom. This model uses a
unit to create the control force between body mass and
wheel mass.
The motion equations of the car body and the wheel are
as follows:


b b a 1 b w s b w
w w a 1 b w 2 w r
m z f k (z z ) c (z z )
m z f k (z z ) k (z z )

+
&& & &
&&
(1)

with the following constants and variables which
respect the static equilibrium position:

m
b
Body mass 250 kg
m
w
Wheel mass 35 kg
k
1
Spring constant (stiffness) of the body 16 000 N/m
k
2
Spring constant (stiffness) of the wheel 160 000
N/m
f
a
Desired force by the cylinder
c
s
Damping ratio of the damper 980 Ns/m
z
r
Road displacements
z
b
Body displacement
z
w
Wheel displacements

To model the road input let us assume that the
vehicle is moving with a constant forward speed. Then
the vertical velocity can be taken as a white noise
process which is approximately true for most of real
roadways.
To transform the motion equations of the quarter
car model into a space state model, the following state
variables are considered:

x = [x
1
, x
2
, x
3
, x
4
]
T
(2)



Fig. 1: One-quarter-car model

where:

x
1
= z
b
body displacement
x
2
=
1 b
x z & & absolute velocity of the body
x
3
= z
w
wheel displacement
x
4
3 w
x z & & absolute velocity of the wheel

Then the motion equations of the quarter car model
for the active suspension can be written in state space
form as follows:


a r
x A.x B.f F.z + + & (3)
With

s s
b b b b b
w
w w w
c c k k
m m m m m
A B F
k
k k k
m
m m m
1 1
1
1 1
1
1 1

1
1 1
1
1 1
1
1 1
1
1 1
+
1
1 1
]
1 1
] ]
1 1
2
1 1 2
0 1 0 0 0
0
1
0
0
0 0 0 0 1
1
0 0
(4)

In general, in the space state a sliding surface is
defined by

s(x,t) kx (5)

where k = [k
1
, k
2
, k
3
, k
4
] in which the k
i
s are all real
and are chosen such that

4
( i 1)
i
i 1
h(r) k r

is a Hurwitz polynomial where r is a Laplace operator.
The control problem will be considered as the state
World Appl. Sci. J., 8 (5): 555-564, 2010
557

vector x remaining on the sliding surface s (x, t) = 0 for
all t0. The sliding mode control process can be
classified into two phases, the approaching phase with s
(x, t) 0 and the sliding phase with s (x, t) = 0 for
initial error x(0)=0. In order to guarantee that the
trajectory of the state vector x will translate from the
approaching phase to the sliding phase, the sufficient
condition

s(x,t) s(x,t) s , 0 > & (6)

must be satisfied. Two type of control law must be
derived separately for those two phases described
above. In the sliding phase, it implies s (x, t) = 0
ands ( x, t ) 0 & . In order to force the system dynamics to
stay on the sliding surface, the equivalent control u
eq

can be derived as follows: If A, B and Z
r
are known,
taking the derivative of the sliding surface with
respective to time, we get


a r
a r
s kx
k(Ax Bf FZ )
kAx kBf kFZ 0

+ +

& &
(7)

f
a
is control input and for simplicity in notation we call
it as u, therefore u=f
a
. Therefore, the equivalent control
can be obtained as


eq r
1
u (kAx k F Z)
kB

+ (8)

Let f (x, t) = kAx + kFZ
r
. On the contrary, in the
approaching phase, s (x, t) 0, a approaching-type
control u
ap
must be added in order satisfy the sufficient
condition (6) and the complete sliding mode control
will be expressed as


eq ap ap
u u u , u sgn(s)

+ (9)

where

<0.
To obtain the sliding mode control (9), the system
matrices A, B, Z
r
and switching parameter

must be
known in advance. However, f (x, t) and

are
unknown, it is impossible to obtain the control (8). The
purpose of this paper is to approximate u
eq
and u
ap
by
interval type-2 fuzzy logic system (FLS). Furthermore,
the adaptive laws to adjust parameters will be derived.

BRIEF DESCRIPTION OF INTERVAL
TYPE-2 FUZZY LOGIC SYSTEM

Due to the complexity of the type reduction, the
general type-2 FLS becomes computationally intensive.



Fig. 2: Interval type-2 fuzzy set with uncertain mean

In order to make things simpler and easier to compute
meet and join operations, the secondary MFs of an
interval type-2 FLS are all unity which leads finally to
simplify type reduction. The 2-D interval type-2
Gaussian membership function (MF) with uncertain
mean m[m
1
, m
2
] and a fixed deviation is shown in
Fig. 2.


2
1 2 A
1 x m
(x) exp , m [ m, m ]
2
1
_
1

,
1
]
%
(10)

It is obvious that the type-2 fuzzy set is in a region
bounded by an upper MF and a lower MF denoted
as
A
(x)
%
and
A
(x)
%
, respectively and is called a Foot of
Uncertainty (FOU). In the meantime, the firing
strength F
i
for the ith rule can be an interval type-1 set
expressed as:


i
i
i
f
F
f
1
1
1 ]
(11)
where


i i i
1 n j
i n
1 n j 1 j
F F F
f (x ) * * (x ) (x )

% % %
L (12)


i i i
1 n j
i n
1 n j 1 j F F F
f (x ) * * (x ) (x )

% % %
L (13)

Based on the center of sets type reduction, the
defuzzified crisp output from an interval type-2 FLS is
the average of y
l
and y
r
, i.e.,


l r
y y
y (x)
2
+
(14)
World Appl. Sci. J., 8 (5): 555-564, 2010
558



Fig. 3: The structure of Type-2 fuzzy logic system

where y
l
and y
r
are the left most and right most points
of the interval type-1 set which can be obtained as:


M L M
i
i i i i i
l l l l
i i 1 i L 1
l M L M
i
i i
l
i i 1 i L 1
f w f w f w
y
f f f
+
+
+

+


(15)
and

M R M
i
i i i i i
r r r r
i i 1 i R 1
r M R M
i
i i
r
i i 1 i R 1
f w f w f w
y
f f f
+
+
+

+


(16)

where M is the total number of rules in the rule base of
the Type-2 FNN. The weighting factors
i
l
w and
i
r
w of
the consequent part represents the centroid interval set
of the consequent type-2 fuzzy set of the i
th
rule. In the
meantime, R and L can be determined by using the
iterative Karnik-Mendel procedure.
A type-2 fuzzy logic system (FLS) is very similar
to a type-1 FLS as shown in Fig. 3,

the major structure
difference being that the defuzzifier block of a type-1
FLS is replaced by the output processing block in a typ-
2 FLS which consists of type-reduction followed by
defuzzification.
There are five main parts in a type-2 FLS: fuzzifier,
rule base, inference engine, type-reducer and
defuzzifier. A type-2 FLS is a mapping f:
p

1
.
After defuzzification, fuzzy inference, type-reduction
and defuzzification, a crisp output can be obtained.
Consider a type-2 FLS having p inputs
x
1
X
1
,, x
p
X
p
and one output yY. The type-2
fuzzy rule base consists of a collection of IF-THEN
rules. As in the type-1 case, we assume there are M
rules and the rule of a type-2 relation between the
input space X
1
X
2
X
p
and the output space Y can
be expressed as:

l l l
1 1 p p
l
R : IF x is F and and x i s F,
THEN y is G l 1,2, , M
% %
L
%
L
(17)

where
l
j
F
% s are antecedent type-2 sets (j = 1,2,,p) and
l
G s
%
are consequent type-2 sets.
The inference engine combines rules and gives a
mapping from input type-2 fuzzy sets to output type-2
fuzzy sets. To achieve this process, we have to
compute unions and intersection of type-2 sets, as
well as compositions of type-2 relations. The output
of inference engine block is a type-2 set. By using
the extension principle of type-1 defuzzification
method, type-reduction takes us from type-2 output
sets of the FLS to a type-1 set called the type-reduced
set. This set may then be defuzzified to obtain a single
cris p value.

CAR SUSPENSION CONTROL BY INDIRECT
ADAPTIVE INTERVAL TYPE-2 FUZZY
SLIDING MODE CONTROL

To begin with, our objective is to use interval type-
2 FNN as shown in Fig. 4 to approximate the nonlinear
functions f (x, t). Based on the Lyapunov approach, the
adaptive laws can be developed to adjust the parameters
of FNN to attenuate the approximation error and
external disturbance.
We replace f (x, t) and u
ap
in Eq. (9) by the interval
type-2 FNN
f
f ( x| ) and
h
h s ( | ) , respectively, as in
Eq. (14)


f r T T T
fr f l f f f
f l
1
f ( y| )
2
1
1

1
]

]
(18)


T T T hr
h h
hr hl h
hl
1
h(s| ) [ ]
2
1

1

]
(19)
World Appl. Sci. J., 8 (5): 555-564, 2010
559



Fig. 4: The structure of interval type-2 fuzzy neural
network

Therefore, if the control input is chosen

( )
f h
1
u f ( x| ) h(s| )
kB
+ (20)

To be assumed that

max h
h(s| ) +

Where
max
is the maximum error which will be
defined in relation (23).
In order to adjust the parameters in the fuzzy logic
systems, we have to derive adaptive laws. Hence, the
optimal parameter estimations
*
f
and
*
h
are defined as


f f
x
*
f f
x

=arg min s up| f ( x| ) f ( x, t ) |

1

1
]
(21)


h h
s
*
ap h h
s

= arg min sup| h(s| ) u |

1

1
]
(22)

where
f
,,
h
,
x
and
s
are constraint sets of suitable
bounds on f, g, h, y x and s, respectively and they
are defined as:

{ } f f f f
| M ,
{ } h h h h
| M
{ }
x x
x | x M
and
{ } s f f f
| M

where M
f
, M
h
, M
x
and M
s
are positive constants.
Define the minimum approximation errors as:


( )
*
f
f(x,t) f ( x| ) (23)

The time derivative of the sliding surface is :


( )
( )
( )
r
*
f h f
* * *
f h h
*
f f
* *
h h h
s kAx kBu kFZ f(x, t) kBu
f ( x| ) f(x, t) h(s| ) f ( x| )
f ( x| ) h(s| ) h(s| )
f ( x| ) f ( x| )
h(s| ) h(s| ) h(s| )
+ +
+
+
+

&
(24)

where
*
f f f

%
and
*
h h h

%

Therefore, from proceeding consideration, the
following theorem can be obtained.

Theorem 1: Consider the car suspension system (1),
the fuzzy-based adaptive laws are chosen as


f f
r s (y)
&
(25)


h h
r s (s)
&
(26)

Then, the overall adaptive scheme guarantees the
global stability of the resulting closed-loop system in
the sense that all signals involved are uniformly
bounded and the tracking error will converge to zero
asymptotically.

Proof: In order to analyze the closed-loop stability, the
Lyapunov function candidate is chosen as


T T
2
f f h h
f h
1 1 1
V s
2 2r 2r
+ +
% % % %
(27)

Taking the derivative of the Eq. (27) with respect
to time, we get


T T
f f h h
f h
1 1
V ss
r r
+ +
& &
% % % % &
& (28)

By substituting (25) and (26) into (28) and using
(24) yields


{ }
T T
*
f h h f f h h
f h
T T
*
f h h f f h h
f h
T T
*
f h f f h h h
f h
T
f f f
f
1 1
V s (x) (s) h(s| )
r r
1 1
s s (x) s (s) sh(s| )
r r
1 1
s r s ( x ) r s ( s ) sh(s| )
r r
1
r s ( y)
r
+ + +
+ + +
_ _
+ + +
, ,
_
+

,
& &
% % % % % % &
& &
% % % % % %
& &
% % % %
&
% %
T
h h h
h
T T
f h f f h h
f h
1
r s (s) s sgn(s)
r
1 1
r s ( y ) r s ( s ) s
r r

_
+

,
_ _
< + +

, ,
&
% %
& &
% % % %
(29)
World Appl. Sci. J., 8 (5): 555-564, 2010
560

where
f f

&
% &
and
h h

&
% &
. Substitute (27)-(29), we
have

V s sgn(s) < s

<
&
(30)

Since k = [k
1
, k
2
, k
3
, k
4
] and x = [x
1
, x
2
, x
3
, x
4
] in which
the kis are all real and are chosen such that

n
( i 1)
i
i 1
h(r) k r

is a Hurwitz polynomial where r is a Laplace operator,
we have

t
limx(t) 0

and
t
lim| s(x,t)| 0

The proof is completed.

SIMULATION EXAMPLE

In this section, we will apply both indirect adaptive
interval type-2 FNN sliding mode controller and type-1
sliding mode controller to evaluate the regulation
performance for a quarter-car suspension system as
follows.
The motion equations of the car body and the
wheel are given in (4). The design procedure is given in
the following steps:

[Step 1]: The feedback gain matrix is chosen as
k = [k
1
, k
2
, k
3
, k
4
] = [3 3 3 1] and the sliding
surface is obtained as s = -kx.
[Step 2]: Specify the design parameters r
f
= 30, r
g
= 1, r
h

= 30, in adaptive laws (27)-(29), simulation
time t
f
=20 second and the step size h =0.01.
[Step 3]: The following type-1 and interval type-2 fuzzy
membership functions for x
i
and y
i
i = 1, 2 are

selected as
j
i
F , j=1,,7 shown in Table 1.
Also, the footprint of uncertainty of the type-2
membership function for x
i
and y
i

i = 1, 2 x
i
,
are as shown in Fig. 5.
[Step 4]: The control law is expressed as

( )
a f h
1
u f f ( x| ) h(s| )
kB
+

The road displacement (external disturbance) is
given as:

r
0.05sin8 t , 0 t 0.25sec, 0.5 t 0.75sec
z( t )
0 otherwise

'

In addition, the simulation results are classified into
two cases as follows: type-1 FNN sliding mode control
and interval type-2 sliding mode control and both noise
free case and training data corrupted with SNR=10 dB
internal noise case are included.

Case 1: Type-1 FNN sliding mode control:

The time responses of the vehicle body bounce and
vertical acceleration of the vehicle body are presented
in Fig. 6. Figure 7 shows the control effort.

-15 -10 -5 0 5 10 15
0
0.2
0.4
0.6
0.8
1


Fig. 5: The interval type-2 Gaussian membership
functions for x
i
and y
i
i = 1, 2

Table 1: Interval type-2 and type-1 fuzzy membership functions for xi and yi i = 1, 2
Mean (m) Mean (m)
---------------------------------------------- ------------------------------------------
Variance () m1 m2 m (type-1) Variance ()
1
i
i F
( y ) 1 -2.5 -1.5 -2.0 5
i
i F
( y ) 1 0.5 1.5 1.0
2
i
i F
( y ) 1 -2.0 -1.0 -1.5 6
i
i F
( y ) 1 1.0 2.0 1.5
3
i
i F
( y ) 1 -1.5 -0.5 -1.0 7
i
i F
( y ) 1 1.5 2.5 2.0
4
i
i F
( y ) 1 -0.5 0.5 0.0
World Appl. Sci. J., 8 (5): 555-564, 2010
561

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.05
0
0.05
xd=0


xd
x1
zr
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.5
0
0.5


xd
d
ot
x2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.1
0
0.1


xd
2
x3
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-2
0
2


xd
2d
ot
x4


Fig. 6: The time responses of the vehicle body bounce and vertical acceleration of the vehicle body

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-1.5
-1
-0.5
0
0.5
1
1.5
2


fa


Fig. 7: Control input u (t)

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.05
0
0.05
MIMO Indirect Adaptive Fuzzy Sliding Mode control-"quarter car"


xd
x
zr
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.5
0
0.5


xd
d
ot
x2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.1
0
0.1


xd
2
x3
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-2
0
2


xd
2d
ot
x4


Fig. 8: The time responses of the vehicle body bounce and vertical acceleration of the vehicle body with training
data corrupted by SNR=10 dB
World Appl. Sci. J., 8 (5): 555-564, 2010
562

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-1.5
-1
-0.5
0
0.5
1
1.5


fa


Fig. 9: Control input u (t)

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.05
0
0.05
xd=0


xd
x1
zr
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.5
0
0.5


xd
d
ot
x2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.1
0
0.1


xd
2
x3
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-2
0
2


xd
2d
ot
x4


Fig. 10: The time responses of the vehicle body bounce and vertical acceleration of the vehicle body

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-2
-1.5
-1
-0.5
0
0.5
1


fa


Fig. 11: Control input u (t)
World Appl. Sci. J., 8 (5): 555-564, 2010
563

Table 2: Comparison of control efforts and regulation errors of type-1 and interval type-2 controllers
Result table ( )
2
1 d
0
x x dt

( )
2
3 d
0
x x dt

Sum fa
Noisefree Type-1 0.01553 0.075174 67.8952
Type-2 0.01547 0.075240 63.2730
Noise SNR=10dB Type-1 0.01557 0.075177 72.3059
Type-2 0.01548 0.075210 63.8285

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.05
0
0.05


xd
x1
zr
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.5
0
0.5


xddot
x2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-0.1
0
0.1


xd
2
x3
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-2
0
2


xd2dot
x4


Fig. 12: The time responses of the vehicle body bounce and vertical acceleration of the vehicle body with training
data corrupted by SNR=10 dB

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-2
-1.5
-1
-0.5
0
0.5
1
1.5


fa


Fig. 13: Control input u (t)

The time responses of the vehicle body bounce and
vertical acceleration of the vehicle body with training
data corrupted by SNR=10 dB are shown in Fig. 8 and
control input is given in Fig. 9.

Case 2: Interval type-2 FNN sliding mode control:

The time responses of the vehicle body bounce and
vertical acceleration of the vehicle body are presented
in Fig. 10. Figure 11 shows the control effort.
The time responses of the vehicle body bounce and
vertical acceleration of the vehicle body with training
data corrupted by SNR=10 dB are shown in Fig. 12 and
control input is given in Fig. 13.
Table 2 manifests that the proposed interval type-2
fuzzy controller gives the better performance compared
with type-1 fuzzy controller as system internal
uncertainties (noisy training data) and external
disturbance appeared. Moreover, the adaptive type-1
fuzzy controller must expend more control effort to deal
with noisy training data.
From above simulation results, we can see that, the
interval type-2 fuzzy sliding mode control can handle
control of car suspension system, very well. It is clear
from Table 2, that the regulation performance of the
interval type-2 FNN sliding mode controller is better
than the performance of the type-1 FNN sliding mode
controller.
World Appl. Sci. J., 8 (5): 555-564, 2010
564

CONCLUSION

A new stable indirect adaptive interval t ype-2 FNN
sliding mode controller is developed to control car
suspension systems. Based on the Laypunov stability
criterion, the asymptotic stability of the whole system
can be achieved and the free parameters of the adaptive
fuzzy sliding mode controller can be tuned on-line by
an output feedback control law and adaptive laws.
Moreover, the chattering phenomena in the control
efforts can be reduced. From the simulation results, it is
obvious that the regulation performance obtained from
interval type-2 fuzzy controller is better than the
regulation performance obtained from type-1 fuzzy
controller with data corrupted with internal noise.

REFERENCES

1. Hrovat, D. , 1987. Ride improvements with
unsprung mass dynamic absorbers for active
suspension vehicles. Research Report SR-87-120
Ford Motor Company.
2. Thompson, A.G. and B.R. Davis, 1988. Optimal
linear active suspensions with derivative
constraints and output feedback control. Veh.
System Dynamics, 17: 179-192.
3. Hrovat, D., 1997. Survey of advanced suspension
developments and related optimal control
applications. Automatica, 33 (10): 1781-1817.
4. Nagai, M., 1993. Recent researches on active
suspensions for ground vehicles. JSME Int. J., Ser.
C, 36 (2): 161-170.
5. Navid Noroozi, Mehdi Roopaei and M. Zolghadri
Jahromi, 2009. Adaptive fuzzy sliding mode
control scheme for uncertain systems
Communications in Nonlinear Science and
Numerical Simulation, 14 (11): 3978-3992.
6. Navid Noroozi, Mehdi Roopaei, Paknosh
Karimaghaee and Ali Akbar Safavi, 2010. Simple
adaptive variable structure control for unknown
chaotic systems Communications in Nonlinear
Science and Numerical Simulation, 15 (3):
707-727.
7. Mehdi Roopaei, Mansoor Zolghadri and Sina
Meshksar, 2009. Enhanced adaptive fuzzy sliding
mode control for uncertain nonlinear systems
Communications in Nonlinear Science and
Numerical Simulation, 14 (9): 3670-3681.
8. Mehdi Roopaei, Mansoor Zolghadri, Robert John
and Tsung-Chih Lin, 2009. Unknown Nonlinear
Chaotic Gyros Synchronization Using Adaptive
Fuzzy Sliding Mode Control with Unknown
Dead-Zone Input Communications in Nonlinear
Science and Numerical Simulation. In Press,
Accepted Manuscript.

9. Mehdi Roopaei, Mansoor Zolghadri Jahromi and
Shahram Jafari, 2009. Adaptive gain fuzzy sliding
mode control for the synchronization of nonlinear
chaotic gyros, Chaos 19 013125.
10. Mehdi Roopaei and Mansoor Zolghadri Jahromi ,
2008. Synchronization of two different chaotic
systems using novel adaptive fuzzy sliding mode
control, Chaos 18 033133.
11. Navid Noroozi, Mehdi Roopaei, Valentina Emilia
Balas and Tsung-Chih Lin, 2009. Observer-based
adaptive variable structure control and
synchronization of unknown chaotic systems .
Applied Computational Intelligence and
Informatics. SACI '09. 5th International
Symposium, Timisoara, Romania, pp: 71-76.
12. Tsung-ChihLin, Han-LeihLiu and Ming-Jen Kuo,
2009. Direct adaptive interval type-2 fuzzy
control of multivariable nonlinear systems .
Engineering Applications of Artificial Intelligence,
22: 420-430.
13. Tsung-Chih Lin, Ming-Jen Kuo and Chao-Hsing
Hsu, 2009. Robust Adaptive Tracking Control of
Multivariable Nonlinear Systems Based on Interval
Type-2 Fuzzy approach. International Journal of
Innovative Computing, Information and Control,
accepted to be published.
14. Chen, S.T., S.L. Tzuu and C.K. Fan, 1995. Design
of fuzzy controlles for active suspension system.
Mechatronics, 5 (4): 365-383.
15. Yun, Q.Z., Y.S. Zhao and H.Z. Yang, 2007. A
dynamic sliding-mode controller with fuzzy
adaptive tuning for an active suspension system.
Proc. IMechE, Part D: J. Automobile Engineering,
221: 417-428.
16. Rao, M.V.C. and V.A. Prahlad, 1997. A tunable
fuzzy logic controller for vehicle active suspension
system. Fuzzy Sets Systems, 85: 11-21.
17. Fernando, J. and D.E. Viassolo, 2000. Fuzzy
control for active suspensions. Mechatronics,
10: 897-920.
18. Huang, S.J. and H.C. Chao, 2000. Fuzzy logic
controller for a vehicle active suspension system.
Proc. Instn Mech. Engrs, Part D: J. Automobile
Engineering, 214: 1-12.
19. Huang, S.J. and W.C. Lin, 2003. Adaptive fuzzy
controller with sliding surface for vehicle
suspension control. IEEE Trans. Fuzzy Systems,
11: 550-559.
20. Lian, R.-J., B.-F. Lin and W.-T. Sie, 2005. Self-
organizing fuzzy control of active suspension
systems. Int. J. Systems Sci., 36: 119-135.
21. Sakman, L.E. and G. Rahmi, 2005. Fuzzy logic
control of vehicle suspensions with dry friction
nonlinearity. Sadhana, 30 (5): 649-659.