International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.

2, April 2012
DOI:10.5121/ijitca.2012.2201 1
ROBUST STABILIZATION OF A QUADROTOR AERIAL
VEHICLE IN PRESENCE OF ACTUATOR FAULTS
Hicham Khebbache
1
, Belkacem Sait
2
, Fouad Yacef
3
and Yassine Soukkou
4
1,2
Automatic Laboratory of Setif (LAS), Electrical Engineering Department,
Setif University, ALGERIA
khebbachehicham@yahoo.fr , sait_belkacem19@yahoo.fr
3,4
Automatic Laboratory of Jijel (LAJ), Automatic Control Department,
Jijel University, ALGERIA
yaceffouad@yahoo.fr , soukkou_yassine@yahoo.fr
ABSTRACT
This paper considers the stabilization problem of an underactuated quadrotor UAV system in presence of
actuator faults. The dynamic model of quadrotor while taking into account various physical phenomena,
which can influence the dynamics of a flying structure is presented. Subsequently, a new control strategy
based on backstepping approach and taking into account the actuator faults is developed. Lyapunov
based stability analysis shows that the proposed control strategy design keeps the stability of the closed
loop dynamics of quadrotor UAV even after the presence of actuator faults. Numerical simulation results
are provided to show the good tracking performance of proposed control laws.
KEYWORDS
Backstepping control, Dynamic model of quadrotor, Fault tolerant control (FTC), Robust control, Sliding
mode control (SMC), Unmanned aerial vehicles (UAV).
1. INTRODUCTION
The automatic control of a quadrotor UAV is not a straight on mainly due to its underactuated
properties. The dynamic model of quadrotor UAV has six degree-of-freedom (DOF) with only
four independent thrust forces generated by four rotors. It is difficult to control all these six
outputs with only four control inputs. Moreover, uncertainties associate with dynamic model
also bring more challenge for control design.
Despite the real progress made, researchers must still deal with serious difficulties, related to the
control of such systems, particularly, in the presence of atmospheric turbulences. In addition,
the navigation problem is complex and requires the perception of an often constrained and
evolutionary environment, especially in the case of low-altitude flights.
In contrast to terrestrial mobile robots, for which it is often possible to limit the model to
kinematics, the control of aerial robots (quadrotors) requires dynamics in order to account for
gravity effects and aerodynamic forces [9].
In [5-10], authors propose a control-law based on the choice of a stabilizing Lyapunov
function ensuring the desired tracking trajectories along (X,Z) axis and roll angle only. The
authors in [1], do not take into account frictions due to the aerodynamic torques nor drag forces.
They propose a control-laws based on backstepping, and on sliding mode control in order to
stabilize the complete system (i.e. translation and orientation). In [12-13], authors take into
account the gyroscopic effects and show that the classical model-independent PD controller can
stabilize asymptotically the attitude of the quadrotor aircraft. Moreover, they used a new
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
2
Lyapunov function, which leads to an exponentially stabilizing controller based upon the PD
2
and the compensation of coriolis and gyroscopic torques. In [8] the authors develop a PID
controller in order to stabilize altitude. The authors in [3] propose a control algorithm based
upon sliding mode using backstepping approach allowed the tracking of the various desired
trajectories expressed in term of the center of mass coordinates along (X,Y,Z) axis and yaw
angle. In [7], authors used a controller design based on backstepping approach. Moreover, they
introduced two neural nets to estimate the aerodynamic components, one for aerodynamic
forces and one for aerodynamic moments. While in [6] the authors propose a hybrid
backstepping control technique and the Frenet-Serret Theory (Backstepping+FST) for attitude
stabilization, that includes estimation of the desired angular acceleration as a function of the
aircraft velocity. The authors in [4] develop a direct adaptive sliding mode control for the
quadrotor attitude stabilization and altitude trajectory tracking, with consideration of
disturbances and parameter uncertainties. However, all authors mentioned above do not take
into account the faults affecting the actuators of quadrotor, which makes these control-laws very
limited and induce undesired behavior of quadrotor, or even to instability of the latter after
occurrence of actuator faults.
In this paper, the stabilization problem of the quadrotor aircraft in presence of actuator faults is
considered. The dynamical model describing the quadrotor aircraft motions and taking into a
account for various parameters which affect the dynamics of a flying structure such as frictions
due to the aerodynamic torques, drag forces along (X,Y,Z) axis and gyroscopic effects is
presented. Subsequently, a new control strategy based on backstepping approach and taking into
account the actuator faults is developed. In order to compensate the effects of actuator faults,
this control strategy uses a "sign" function as compensation technique. Finally all synthesized
control laws are highlighted by simulations which gave fairly satisfactory results despite the
occurrence of actuator faults.
2. DYNAMICAL MODEL
2.1. Quadrotor dynamic model
The quadrotor have four propellers in cross configuration. The two pairs of propellers {1,3} and
{2,4} as described in Fig. 1, turn in opposite directions. By varying the rotor speed, one can
change the lift force and create motion. Thus, increasing or decreasing the four propellers
speeds together generates vertical motion. Changing the 2 and 4 propellers speed conversely
produces roll rotation coupled with lateral motion. Pitch rotation and the corresponding lateral
motion; result from 1 and 3 propellers speed conversely modified. Yaw rotation is more subtle,
as it results from the difference in the counter-torque between each pair of propellers.
Figure 1. Quadrotor configuration
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
3
The quadrotor model (position and orientation dynamic) obtained is given by [2], [3], [4]:
, ,
, ,
, ,
2
2
2
3
2
4
1
1
1
1
1
1
cos( ) cos( )
y z
f ax r
r
x x x x
f ay z x
r
r
y y y y
x y
f az
z z z
f tx
x
f ty
y
f tz
I I
K J l
u
I I I I
K I I
J l
u
I I I I
I I
K
u
I I I
K
x x u u
m m
K
y y u u
m m
K
z z g u
m m
| 0 0 |
0 | | 0
0|
| 0
!

! +
!
!

!
+ +
!
!
!
+ !
,
!
!
+
!
!
!
+
!
!
+ !

(1)
with
, ,
, ,
1 2 3 4
cos cos sin sin sin
cos sin sin sin cos
r
x
y
u
u
e e e e
| 0 |
| 0 |
!
+
!
!
+
,
!

!

(2)
u
1
, u
2
, u
3
and u
4
are the control inputs of the system which are written according to the angular
velocities of the four rotors as follows:
2
1 1
2
2 2
2
3 3
2
4 4
0 0
0 0
u b b b b
u lb lb
u lb lb
u d d d d
e
e
e
e
1 1 1
1 1 1

1 1 1

1 1 1
1 1 1

1
] ] ]
(3)
From (2) it easy to show that :
, , , , , ,
, , , , , ,
, ,
arcsin sin cos
cos sin
arcsin
cos
d x d y d
x d y d
d
d
u u
u u
|

0
|
!

!
!
_
+ ,

!

!
,
(4)
2.2. Rotor dynamic model
The dynamics of a DC motor is given by the following differential equations:
{ ] , 1, 2, 3, 4
r i i i
J Q i e t (5)
With Q
i
=d
i
2
is the reactive torque generated, in free air, by the rotor i due to rotor drag, and

i
is the input torque.
A control law for the input torque
i
is developed in [12-13], it is given by:
, i i r d i i i
Q J k t e e +
(6)
where k
i
, i {1,,4} are four positive parameters,
d,i
, i {1,,4} are the desired speed of
each rotor and
, i i d i
e e e .
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
4
In fact, applying (6) to (5) leads to
i
i i
r
k
J
e e

(7)
which shows the exponential convergence of
i
to
d,i
and hence the convergence of the
airframe torques to the desired values leading to the attitude stabilization of the quadrotor
aircraft.
In our application, the dc motors are voltage controlled, so we need to obtain the voltage input
to each motor. Assuming that the motor inductance is small and taking into consideration the
gear ratio, one can obtain the voltage to be applied to each motor as follows [12-13]:
{ ] , 1, 2, 3, 4
a
i i m g i
m g
R
v k k i
k k
t e + (8)
where R
a
is the motor resistance, k
m
is the motor torque constant, and k
g
is the gear ratio.
3. CONTROL STRATEGY OF QUADROTOR WITH ACTUATOR FAULTS
The complete model resulting by adding the actuator faults in the model (1) can be written in
the state-space form:
, , , , , , , , , , , ,
, ,
a
x t x t g x t u t f t g + + (9)
With
, ,
n
x t is the state vector of the system,
, ,
m
u t is the input control vector, and
( )
q
a
f t is the resultant vector of actuator faults related to the quadrotor motions, such as:
[
1 12
,..., , , , , , , , , , , ,
T
T
X x x x x y y z z | | 0 0 1
]

(10)
From (1), (10) and considering the actuator faults, we obtain the following state representation:
, ,
, ,
, ,
, ,
1 2
2
2 1 4 6 2 2 3 4 1 2 1
3 4
2
4 4 2 6 5 4 6 2 2 3 2
5 6
2
6 7 2 4 8 6 3 4 3
7 8
8 9 8 1
9 10
10 10 10 1
11 12
12 11 12 1 4
1
1
cos( ) cos( )
r a
r a
a
x
y
a
x x
x a x x a x a x b u f
x x
x a x x a x a x b u f
x x
x a x x a x b u f
x x
x a x u u
m
x x
x a x u u
m
x x
x a x g u f
m
| 0
!
!
+ + + +
!
!

!
+ + + +

+ + +

,
+

+ +

!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!

(11)
with
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
5
1 2 3 4 5
6 7 8 9 10
11 1 2 3
1
y z fay
fax r z x
x x x y y
x y fty
faz ftx r
y z z
ftz
x y z
I I K
K J I I
a a a a a
I I I I I
I I K
K K J
a a a a a
I I I m m
K l l
a b b b
m I I I
! _ _
!


! ,
,
!
_
!

,
, !
!
!
!

(12)
The following assumptions is needed for the analysis,
Assumption 1: The resultant of actuator faults related to the quadrotor motions are bounded,
, , [ , i 1, 2,3, 4
ai ai
f t f
+
(13)
where {f
a1
+
, f
a2
+
, f
a3
+
, f
a4
+
} are positive constants.
Assumption 2: The unknowns parts
ai
(x, f
ai
, t) including the resultants of actuator faults related
to the quadrotor motions are also bounded,
, , , , [ , , , , i 1,2,3,4
ai ai i ai ai
x f t g x t f k
+
< (14)
where {k
a1
, k
a2
, k
a3
, k
a4
} are also positive constants.
The adopted control strategy is based on two loops (internal loop and external loop). The
internal loop contains four control laws: control of roll, control of pitch, control of yaw and
control of altitude. The external loop includes two control laws of positions x and y. The
external control loop generates a desired of roll (
d
) and pitch (
d
) through the correction block
(illustrated by equation (4)). This block corrects the rotation of the roll and pitch depending on
the desired yaw (
d
). The synoptic scheme below shows the control strategy:
Figure 2. Synoptic scheme of the proposed control strategy
Control of
law
Control of
pitch
Control of
roll
Control of
altitude Z
Correction
block
Control of
y position
Control of
x position
u
z
u

u
x
u
y

d
y
d
x
d
Z
d

d
Desired
trajectory of
x position
Desired
trajectory of
y position
Desired
trajectory of
Z altitude
Desired
trajectory of
law
Internal loop
External loop
, , ,Z
x,y
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
6
Basing on backstepping approach, a recursive algorithm is used to synthesize the control laws
forcing the system to follow the desired trajectory in presence of actuator faults, we simplify all
stages of calculation concerning the tracking errors and Lyapunov functions in the following
way:
[
[
( 1) ( 1) ( 1)
i 1,3,5,7,9,11
i 2,4,6,8,10,12
i id
i
i i d i i
x x
e
x x c e

!
!

,
+
!

(15)
such as
[ 0 i 1,...,12
i
c > (16)
and
[ [ ( ), i 2, 4, 6,12 and j 1,..., 4
j aj i
k sign e (17)
The corresponding Lyapunov functions are given by:
[
[
2
2
1
1
i 1, 3, 5, 7, 9,11
2
1
i 2, 4, 6, 8,10,12
2
i
i
i i
e
V
V e

!
!

,
!
+
!

(18)
The synthesized stabilizing control laws are as follows:
, ,
, ,
, ,
2
2 1 1 1 2 1 2 2 1 4 6 2 2 3 4 1 2
1
2
3 3 3 3 4 3 4 4 4 2 6 5 4 6 2 2 4
2
2
4 5 5 5 6 5 6 6 7 2 4 8 6 3 6
3
7
1
1
( ) ( )
1
( ) ( )
1
( ) ( )
(
d r a
d r a
d a
x d
u c c e e e c e a x x a x a x k sign e
b
u c c e e e c e a x x a x a x k sign e
b
u c c e e e c e a x x a x k sign e
b
m
u x c c
u
|
0

+
+
+

, ,
, ,
,
7 7 8 7 8 8 9 8 1
9 9 9 10 9 10 10 10 10 1
1
1 11 11 11 12 11 12 12 11 1
1 3
) / 0
( ) / 0
( )
cos( ) cos( )
y d
d
e e e c e a x u
m
u y c c e e e c e a x u
u
m
u z c c e e e c e a x
x x
+
+
+

,
2 4 12
( )
a
g k sign e
!
!
!
!
!
!
!
!
!
,
!
!
!
!
!
!
!
+
!

(19)
Proof
For i =1:
1 1 1
2
1 1
1
2
d
e x x
V e
!
!
,

(20)
and
1 1 1 1 2 1
( )
d
V e e e x x

(21)
By application of Lyapunov theorem, the stabilization of e
1
can be obtained by introducing a
new virtual control x
2
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
7
, ,
2 1 1 1 1 1
0
d
d
x x c e c o > (22)
It results that:
2
1 1 1
0 V c e

(23)
For i =2:
2 2 1 1 1
2
2 1 2
1
2
d
e x x c e
V V e
+ !
!
,
+
!

(24)
and
, , , ,
2
2 1 1 1 2 2 1 4 6 2 2 3 4 1 1 1 1 2 1 1 2
( )
r d a
V e c e e e a x x a x a x x c c e e b u + + + + + + + +

(25)
We know a priori from (14) that:
1 1 1 1 1 1 a a a a
b f b f k
+
< (26)
The stabilization of (e
1
, e
2
) can be obtained by introducing the input control u
2
, ,
2
2 1 1 1 1 2 1 2 2 1 4 6 2 2 3 4 1 2
1
1
( ) 0
d r
u x c c e e e c e a x x a x a x c
b
+ > (27)
It result that
2 2
2 1 1 2 2 2 1 1
( )
a
V c e c e e

(28)
In order to compensate the resultant of actuator faults related to the roll motion (
a1
), a sign
function is introduced. We take:
1 1 2
( )
a
k sign e (29)
The equation (28) is then
2 2
2 1 1 2 2 2 1 1
( ) 0
a a
V c e c e e k

(30)
The same steps are followed to extract u
3
, u
4
, u
x
, u
y
and u
1
.
4. SIMULATION RESULTS
Two cases are treated to evaluate the performances of the controller developed in this paper.
1- Results without faults are shown in Figure. 3, Figure. 5, Figure. 7 and Figure .9.a.
2- Results with four resultants of actuator faults {f
a1
, f
a2
, f
a3
, f
a4
} related to roll, pitch, yaw and
altitude motions with 25% of maximum values of inputs control u
2
, u
3
, u
4
and u
1
at 15s,
20s, 25s and 30s respectively are shown in Figure. 4, Figure. 6, Figure. 8 and Figure .9.b.
The simulation results are obtained based on the real parameters in Table. 1 (see the appendix).
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
8
0 10 20 30 40 50 60
-6
-4
-2
0
2
4
6
x 10
-3
Time[sec]
R
o
ll a
n
g
le
[ra
d
]

0 10 20 30 40 50 60
-6
-4
-2
0
2
4
6
8
x 10
-3
Time[sec]
P
itc
h
a
n
g
le
[ra
d
]

0 10 20 30 40 50 60
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Time[sec]
Y
a
w
a
n
g
le
[ra
d
]

(a) evolution of roll angle (b) evolution of pitch angle (c) evolution of yaw angle
0 10 20 30 40 50 60
0
1
2
3
4
5
6
Time[sec]
X
p
o
s
itio
n
[m
]
X
d
X
0 10 20 30 40 50 60
-3
-2
-1
0
1
2
3
Time[sec]
Y
p
o
s
itio
n
[m
]
Y
d
Y
0 10 20 30 40 50 60
-1
0
1
2
3
4
5
6
Time[sec]
Z
a
ttitu
d
e
[m
]
Z
d
Z
(d) evolution of x position (e) evolution of y position (f) evolution of z position
Figure. 3. Tracking simulation results of trajectories along roll ( ), pitch ( ), yaw angle ( ) and
(X,Y,Z) axis, Test 1.
0 10 20 30 40 50 60
-8
-6
-4
-2
0
2
4
6
8
x 10
-3
Time[sec]
R
o
ll a
n
g
le
[ra
d
]

0 10 20 30 40 50 60
-6
-4
-2
0
2
4
6
8
x 10
-3
Time[sec]
P
itc
h
a
n
g
le
[ra
d
]

0 10 20 30 40 50 60
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Time[sec]
Y
a
w
a
n
g
le
[ra
d
]

(a) evolution of roll angle (b) evolution of pitch angle (c) evolution of yaw angle
0 10 20 30 40 50 60
0
1
2
3
4
5
6
Time[sec]
X
p
o
s
itio
n
[m
]
X
d
X
0 10 20 30 40 50 60
-3
-2
-1
0
1
2
3
Time[sec]
Y
p
o
s
itio
n
[m
]
Y
d
Y
0 10 20 30 40 50 60
-1
0
1
2
3
4
5
6
Time[sec]
Z
a
ttitu
d
e
[m
]
Z
d
Z
(d) evolution of x position (e) evolution of y position (f) evolution of z position
Figure. 4. Tracking simulation results of trajectories along roll ( ), pitch ( ), yaw angle ( ) and
(X,Y,Z) axis, Test 2.
Figure. 3 and Figure. 4 shows a good tracking of the desired trajectories for both Tests, while
seeing a low transient deviations due after occurrence of the resultant of actuator faults related
to altitude motion at 30s (illustrated respectively by (a), (b) and (f) in Test 2), which explains
well the under-actuating of our system.
According to Figure. 5 and Figure. 6, there is also a good tracking of the desired velocities for
both Tests. Moreover, Figure. 6 clearly shows a small transient peaks (illustrated by (a) and (b)),
27.5 28 28.5 29 29.5 30 30.5 31 31.5 32 32.5
-6
-5.5
-5
-4.5
-4
-3.5
-3
x 10-3
Time[sec]
Roll angle[rad]
29.2 29.4 29.6 29.8 30 30.2 30.4 30.6 30.8 31 31.2
-2.2
-2
-1.8
-1.6
-1.4
-1.2
x 10-3
Time[sec]
Pitch angle[rad]
30.5 31 31.5 32 32.5 33
4.2
4.4
4.6
4.8
5
5.2
Time[sec]
Z attitude [m]
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
9
and low deviations (illustrated by (c) and (f)) disappears quickly after appearance of the
resultants of actuator faults related respectively to yaw and altitude motions.
0 10 20 30 40 50 60
-4
-3
-2
-1
0
1
2
3
x 10
-3
Time[sec]
A
n
g
u
la
r v
e
lo
c
ity
o
f ro
ll [ra
d
/s
]

d
p
p
0 10 20 30 40 50 60
-2
-1
0
1
2
3
4
x 10
-3
Time[sec]
A
n
g
u
la
r v
e
lo
c
ity
o
f p
itc
h
[ra
d
/s
]

d
p
p
0 10 20 30 40 50 60
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Time[sec]
A
n
g
u
la
r v
e
lo
c
ity
o
f y
a
w
[ra
d
/s
]

d
p
p
(a) evolution of angular velocity of roll (b) evolution of angular velocity of pitch (c) evolution of angular velocity of yaw
0 10 20 30 40 50 60
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Time[sec]
L
in
e
a
r v
e
lo
c
ity
a
lo
n
g
X
a
x
is
[m
/s
]
x
d
p
x p
0 10 20 30 40 50 60
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Time[sec]
L
in
e
a
r v
e
lo
c
ity
a
lo
n
g
Y
a
x
is
[m
/s
]
y
d
p
y p
0 10 20 30 40 50 60
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Time[sec]
lin
e
a
r v
e
lo
c
ity
o
f Z
a
ttitu
d
e
[m
/s
]
z
d
p
z p
(d) evolution of linear velocity along X
axis
(e) evolution of linear velocity along Y
axis
(f) evolution of linear velocity along Z
axis
Figure. 5. Tracking simulation results of the angular and linear velocities, Test 1.
0 10 20 30 40 50 60
-4
-3
-2
-1
0
1
2
3
x 10
-3
Time[sec]
A
n
g
u
la
r v
e
lo
c
ity
o
f ro
ll [ra
d
/s
]

d
p
p
0 10 20 30 40 50 60
-2
-1
0
1
2
3
4
x 10
-3
Time[sec]
A
n
g
u
la
r v
e
lo
c
ity
o
f p
itc
h
[ra
d
/s
]

d
p
p
0 10 20 30 40 50 60
-0.05
0
0.05
0.1
0.15
0.2
0.25
0.3
Time[sec]
A
n
g
u
la
r v
e
lo
c
ity
o
f y
a
w
[ra
d
/s
]

d
p
p
(a) evolution of angular velocity of roll (b) evolution of angular velocity of pitch (c) evolution of angular velocity of yaw
0 10 20 30 40 50 60
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Time[sec]
L
in
e
a
r v
e
lo
c
ity
a
lo
n
g
X
a
x
is
[m
/s
]
x
d
p
x p
0 10 20 30 40 50 60
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Time[sec]
L
in
e
a
r v
e
lo
c
ity
a
lo
n
g
Y
a
x
is
[m
/s
]
y
d
p
y p
0 10 20 30 40 50 60
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Time[sec]
lin
e
a
r v
e
lo
c
ity
o
f Z
a
ttitu
d
e
[m
/s
]
z
d
p
z p
(d) evolution of linear velocity along X
axis
(e) evolution of linear velocity along Y
axis
(f) evolution of linear velocity along Z
axis
Figure. 6. Tracking simulation results of the angular and linear velocities, Test 2.
From Figures. 7 and 8, it is clear to see the appearance of chattering phenomenon in inputs
control u
2
, u
3
and u
4
for both tests, resulting from the use of discontinuous term "sign". Further-
more, Figure. 8 clearly shows a variation of the evolution of inputs control u
2
and u
3
(illustrated
by (b) and (c)) after 15s and 20s. Consequently, the chattering phenomenon in these inputs
control become more intensive. Otherwise, we see well a considerable deviation in evolution of
inputs control u
4
and u
1
(illustrated by (d) and (a)) after 25s and 30s.
30 31 32 33 34 35 36 37 38
-0.38
-0.36
-0.34
-0.32
-0.3
-0.28
-0.26
-0.24
-0.22
Time[sec]
linear velocity of Z attitude [m/s]
25 26 27 28 29 30 31 32 33 34 35
-2
-1
0
1
2
3
4 x 10-4
Time[sec]
Angular velocity of yaw [rad/s]
22 24 26 28 30 32 34 36
-10
-5
0
5
x 10-4
Time[sec]
Angular velocity of roll [rad/s]
27 28 29 30 31 32 33
1
2
3
4
5
6
7
8
9
10
x 10-4
Time[sec]
Angular velocity of pitch [rad/s]
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
10
0 10 20 30 40 50 60
4.6
4.65
4.7
4.75
4.8
4.85
4.9
4.95
5
5.05
5.1
Time[sec]
In
p
u
t c
o
n
tro
l u
1
0 10 20 30 40 50 60
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
x 10
-4
Time[sec]
In
p
u
t c
o
n
tro
l u
2
(a) evolution of input control u1 (b) evolution of input control u2
0 10 20 30 40 50 60
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
x 10
-4
Time[sec]
In
p
u
t c
o
n
tro
l u
3
0 10 20 30 40 50 60
-4
-2
0
2
4
6
8
10
12
x 10
-4
Time[sec]
In
p
u
t c
o
n
tro
l u
4
(c) evolution of input control u3 (d) evolution of input control u4
Figure. 7. Simulation results of all inputs control, Test 1.
0 10 20 30 40 50 60
3.6
3.8
4
4.2
4.4
4.6
4.8
5
5.2
Time[sec]
In
p
u
t c
o
n
tro
l u
1
0 10 20 30 40 50 60
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
x 10
-4
Time[sec]
In
p
u
t c
o
n
tro
l u
2
(a) evolution of input control u1 (b) evolution of input control u2
0 10 20 30 40 50 60
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
2.5
x 10
-4
Time[sec]
In
p
u
t c
o
n
tro
l u
3
0 10 20 30 40 50 60
-4
-2
0
2
4
6
8
10
12
x 10
-4
Time[sec]
In
p
u
t c
o
n
tro
l u
4
(c) evolution of input control u3 (d) evolution of input control u4
Figure. 8. Simulation results of all inputs control, Test 2.
All these changes in the evolution of inputs control u
1
, u
2
, u
3
and u
4
are resulted from the
compensation of the resultants of actuator faults related to Roll, pitch, yaw, and altitude
motions, whose objective is to ensure the stability of the closed loop dynamics of quadrotor
after the occurrence of these actuator faults.
The simulation results given by Figure. 9 clearly shows a good performances and robustness
towards stability and tracking even after the occurrence of the resultants of actuator faults
related to Roll, pitch, yaw, and altitude motions, which explains the efficiency of control
strategy developed in this paper.
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
11
0
2
4
6
-4
-2
0
2
4
-2
0
2
4
6
Xcoordinate [m]
3D position
Y coordinate [m]
Z

c
o
o
r
d
i
n
a
t
e

[
m
]
Reference trajectory
Real trajectory
0
2
4
6
-4
-2
0
2
4
-2
0
2
4
6
Xcoordinate [m]
3D position
Y coordinate [m]
Z

c
o
o
r
d
i
n
a
t
e

[
m
]
Reference trajectory
Real trajectory
(a) evolution of position along (X,Y,Z) axis, Test 1 (b) evolution of position along (X,Y,Z) axis, Test 2
Figure. 9. Global trajectory of quadrotor in 3D.
5. CONCLUSIONS AND FUTUR WORKS
In this paper, we proposed a new control strategy based on backstepping approach and including
the actuator faults. Firstly, we presented the dynamical model of quadrotor taking into account
the different physics phenomena which can influence the evolution of our system in the space.
Secondly, we synthesized a stabilizing control laws in the presence of actuator faults. The
simulation results have shown high efficiency of this control Strategy, it keeps the stability and
the performances of quadrotor during a malfunction of these actuators. As prospects we hope to
develop other control strategies in order to eliminate the chattering phenomenon in inputs
control, while maintaining the stability and the performances of this system, with implement-
ation them on a real system.
6. APPENDIX
Table 1. Quadrotor parameters.
Parameter Value
m 0.486 kg
g 9.806 m/s
2
l 0.25 m
b 2.9842 10
5
N/rad/s
d 3.2320 10
7
N.m/rad/s
J
r 2.8385 10
5
kg.m
2
I
x 3.8278 10
3
kg.m
2
I
y 3.8278 10
3
kg.m
2
I
z 7.1345 10
3
kg.m
2
K
fax 5.5670 10
4
N/rad/s
K
fay 5.5670 10
4
N/rad/s
K
faz 6.3540 10
4
N/rad/s
K
ftx
0.0320 N/m/s
K
fty
0.0320 N/m/s
K
ftz
0.0480 N/m/s
k
m 4.3 10
3
N.m/A
k
g
5.6
R
a
0.67
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
12
7. REFERENCES
[1] S. Bouabdellah, & R.Siegwart. (2005) Backstepping and sliding mode techniques applied to an
indoor micro quadrotor, Proceeding of the IEEE, ICRA, Barcelona, Spain, pp. 2259-2264.
[2] H. Bouadi, M. Bouchoucha, & M. Tadjine. (2007) Modelling and Stabilizing Control Laws
Design Based on Backstepping for an UAV Type-Quadrotor, Proceeding of 6th IFAC
Symposium on IAV, Toulouse, France.
[3] H. Bouadi, M. Bouchoucha, & M. Tadjine. (2007) Sliding Mode Control Based on Backstepping
Approach for an UAV Type-Quadrotor, International Journal of Applied Mathematics and
Computer Sciences, Vol. 4, No. 1, pp. 12-17.
[4] H .Bouadi, S. Simoes Cunha, A. Drouin, & F. Mora-Camino (2011) Adaptive Sliding Mode
Control for Quadrotor Attitude Stabilization and Altitude Tracking Proceeding of 12th IEEE
International Symposium on Computational Intelligence and Informatics, Budapest, Hungary, pp.
449-455.
[5] P. Castillo, A. Dzul, & R. Lozano. (2004) Real-Time Stabilization and Tracking of a Four-Rotor
Mini Rotorcraft, IEEE Transactions on Control Systems Technology, Vol. 12, No. 4, pp. 510-
516.
[6] J. Colorado, A. Barrientos, , A. Martinez, B. Lafaverges, & J. Valente. (2010) Mini-quadrotor
attitude control based on Hybrid Backstepping & Frenet-Serret Theory, Proceeding of the IEEE,
ICRA, Anchorage, Alaska, USA, pp. 1617-1622.
[7] A. Das, F. Lewis, K. Subbarao (2009) Backstepping Approach for Controlling a Quadrotor
Using Lagrange Form Dynamics. Journal of Intelligent and Robotic Systems, Vol. 56, No. 1-2,
pp. 127-151.
[8] L. Derafa, T. Madani, & A. Benallegue. (2006) Dynamic modelling and experimental
identification of four rotor helicopter parameters, Proceeding of the IEEE-ICIT , Mumbai, India,
pp. 1834-1839.
[9] N. Guenard, T. Hamel, & R. Mahony. (2008) A Practical Visual Servo Control for an Unmanned
Aerial Vehicle, IEEE Transactions on Robotics, Vol. 24, No. 2, pp. 331-340.
[10] R. Lozano, P. Castillo, & A. Dzul. (2004) Global stabilization of the PVTOL: real time
application to a mini aircraft, International Journal of Control, Vol. 77, No. 8, pp. 735-740.
[11] J.J.E. Slotine, & W. Li. (1991) Applied nonlinear control, Prentice Hall, Englewood Cliffs, NJ.
[12] A. Tayebi, & S. Mcgilvray. (2004) Attitude stabilisation of a four rotor aerial robot, Proceeding
of the IEEE Conference on Decision and Control, Atlantis Paradise Island, Bahamas, pp. 1216-
1217.
[13] A. Tayebi, & S. McGilvray (2006) Attitude stabilization of a VTOL Quadrotor Aircaft, IEEE
Transactions on Control Systems Technology, Vol. 14, No. 3, pp. 562-571.
Authors
Hicham KHEBBACHE is a Graduate student (Magister) of Automatic Control at the
Electrical Engineering Department of Setif University, Algeria. He received the
Engineer degree in Automatic Control from Jijel University, Algeria, in 2009. He is
with Automatic Laboratory of Setif (LAS). His research interests include Aerial
robotics, Linear and Nonlinear control, Robust control, Fault tolerant control (FTC),
Diagnosis, Fault detection and isolation (FDI).
Belkacem SAIT is an Associate Professor at Setif University and member at
Automatic Laboratory of Setif (LAS), Algeria. He received the Engineer degree in
Electrical Engineering from National Polytechnic school of Algiers (ENP) in 1987, the
Magister degree in Instrumentation and Control from HCR of Algiers in 1992, and the
Ph.D. in Automatic Product from Setif University in 2007. His research interested
include discrete event systems, hybrid systems, Petri nets, Fault tolerant control (FTC),
Diagnosis, Fault detection and isolation (FDI).
International Journal of Information Technology, Control and Automation (IJITCA) Vol.2, No.2, April 2012
13
Fouad YACEF is currently a Ph.D. student at the Automatic Control Department of
Jijel University, Algeria. He received the Engineer degree in Automatic Control from
Jijel University, Algeria, in 2009, and the Magister degree in Control and Command
from Military Polytechnic School (EMP), Algiers, in November 2011. He is with
Automatic Laboratory of Jijel (LAJ). His research interests include Aerial robotics,
Linear and Nonlinear control, LMI optimisation, Analysis and design of intelligent
control systems.
Yassine SOUKKOU is currently a Graduate student (Magister) of Automatic Control
and signal processing at the Automatic Control Department of Jijel University, Algeria.
He received the Engineer degree in Automatic Control from Jijel University, Algeria,
in 2009. His research interests include Image and Signal processing, Robotics, Linear
and Nonlinear control, Fault tolerant control (FTC), Diagnosis, Fault detection and
isolation (FDI).