Applied Mechanics and Materials Vols.

88-89 (2011) pp 88-92

Online available since 2011/Aug/22 at www.scientific.net
© (2011) Trans Tech Publications, Switzerland
doi:10.4028/www.scientific.net/AMM.88-89.88

Adaptive Control of Uncertain Gun Control System of Tank

Lujuan Shen, Ye Bao, Jianping Cai
Zhejiang Water Conservancy And Hydropower College,Hangzhou, Zhejiang, China
shenlj, baoye, caijp@zjwchc.com

Keywords: adaptive control; backlash; nonlinear control; gun control system

Abstract. In this paper, a class of gun control system of tank is considered with uncertain parameters
and the backlash-like hysteresis which modeled by a differential equation. An adaptive control law
is designed with backstepping technique. Compared to exist results on tank gun control problem , in
our control scheme, the effect of backlash hysteresis is considered completely than to be linearized
simply and no knowledge is assumed on the uncertain parameters. the stability of closed loop system
and the tracking performance can be guaranteed by this control law. Simulation studies show that this
controller is effective.

Introduction
In the future, with the development of advanced technology, the performance of weapon have been
raised to a higher requirements in war. Gun control system of tank have been deeply studied to
improve the operational performance of tank gun. In recent ten years, many methods have been
developed to control gun system of tank based on the development of control technology for example
in [1]-[3]. In [1] the technique of variable structure control for sliding mode of tank stabilizator is
studied. In [2] and [3], the same tank gun control system as in [1] is considered. In [2], the
performances of the gun control system in the presence of nonlinear friction and backlash is analyzed.
In [3], a new method of adaptive sliding mode robustness control for its structure perturbation and
disturbance is presented. In this note, the same model of tank gun control system shown in [1] is
considered with uncertain parameters and the backlash-like hysteresis which modeled by a
differential function as shown in [4]. With backstepping technology, an adaptive control scheme is
proposed to guarantee the stability of closed loop system and realize the tracking performance.
Compared to exist results on tank gun control problem , in our control scheme, the effect of backlash
hysteresis is considered completely than to be linearized simply and no knowledge is assumed on the
uncertain parameters.
The rest of the paper is outlined as follows. First of all, problem is formulated followed by
presenting system and backlash-like model. Following, control scheme together with systems
performance analysis are given and simulation studies can illustrate the effectiveness of the adaptive
control scheme. Finally, the paper is concluded.

Problem Statement
Similar in [1], we consider the same gun control system of tank, The mathematical model of system is
given as follows
1
x
1 = x2
i
(1)

x2 = − K e KT x2 + KKT u + 1 d1 (t )
JRa JRa J
where J present the moment of inertia, KT is torque coefficient of motor actuator, K e is the
R
back-EMF coefficient, a is armature-circuit resistance, i presents reduction ratio and K is the
amplifier gain. d (t ) is unknown external disturbance with unknown constant bound. x1 , x2 ∈ R are
the states of systems and denote output angular displacement and motor angular velocity.

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP,
www.ttp.net. (ID: 128.178.131.113, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland-22/03/15,14:36:28)
 Applied Mechanics and Materials Vols. 88-89 89

In order to simplify the system model, the following coordinate transformation are introduced
1
x1 = x1 ; x2 = x2 (2)
i
With above coordinate transformation, the system function can be rewritten as follows
x
1 = x2
(3)
x
2 = ax2 + bu + d1 (t )
where
KK KKT 1
a=− e T , b = , d1 (t ) = d (t )
JRa JRa i Ji
K K R
In system (1), we assume the parameter i is known and other parameters J , T , e , a are
d (t )
unknown. Because J is constant parameter, so 1 can be seen as external disturbance with
unknown bound.
Note that the gun control system of tank is proceed by motor actuator, so the backlash
nonlinearities can not be neglected. We consider the backlash-like input in here and the mathematical
model of backlash-like similar to [ 4, 5] is given as
du (t ) dv dv
=α (cv − u ) + B1 (c > B1 > 0) (4)
dt dt dt
where
α , c, B1 are unknown constant parameters which can be estimated in our adaptive control
scheme. According to the analysis in [4], this differential equation can be solved as
u (t ) = cv(t ) + d 2 (v)
v
−α ( v − v0 )sgn( v
) −α v sgn( v
) ας sgn( v
)
d 2 (v) = [u0 − cv0 ]e +e ∫ [ B − c]e
v0
1 dς (5)

d (v )
As shown in [4], the term 2 is bounded. Considering the backlash-like hysteresis input, the
system (3) can be rewritten as follows
x
1 = x2
(6)
x
2 = ax2 + bcv + d1 (t ) + d 2 (v)
d (t ) + d 2 (v)
Similar in [5], the bounded term 1 can be seen as the ‘disturbance-like’ term denoted by
d (t ) , then system (6) can be written as follows
x
1 = x2
(7)
x
2 = ax2 + bv + d (t )
where b = bc and a are unknown parameters. d (t ) is disturbance-like term with unknown
bound.

Design of Adaptive Controllers

Our control purpose is to design state feedback adaptive control law and update laws to guarantee the
closed loop system globally stable and achieve asymptotic tracking performance, namely the system
y = x1 y (t )
output can asymptotically track the reference signal r .
To design adaptive controller, the following Assumptions are necessary.
Assumption 1: The unknown parameters b = bc is such that sign(b) is known.
Assumption 2: Reference signal yr (t ) and it's ith (i = 1, 2) derivatives are known and bounded.
The following change of coordinates are introduced for designing adaptive control scheme.
Variable z1 express tracking error.
90 Computer-Aided Design, Manufacturing, Modeling and Simulation

z1 = x1 − yr ; z2 = x2 − α1 − y
r (8)
where α1 is the virtual control in step 1. Following we will give the details of adaptive control
scheme designation with backstepping technology. Note that the gun control system of tank is 2
order system, so there are two steps in our control design.
z
Step 1: From (7) and (8), the derivative of 1 can be written as follows
z
1 = x
1 − y
r = z2 + α1 + y
r − y
r = z2 + α1 (9)
The virtual control
α1 in this step can be chosen as follows
α1 = −k1 z1 (10)
k1 > 0 is a positive constant. The Lyapunov function can be chosen as
where
1
V1 = z12
2 (11)
In process of backstepping designation, The positive function showing in above is not the final
V
Lyapunov function. The derivation of 1 is as follows
V
1 = z1 z
1 = z1 z2 − k1 z12 (12)
z
Step 2:From (8) and system function, the derivative of 2 can be written as follows
z
2 = x
2 − α
1 −


yr = ax2 + bv + d (t ) − α
1 −

yr (13)
Note that the gain of input signal v in system model (7) is unknown, so we must design an
estimator to compensate this uncertain parameters. Control input v can be designed as
v = ρˆ v ; v = − z1 − ax yr − sgn( z2 ) Dˆ
ˆ 2 − k2 z2 + α
1 +


(14)
1
k >0
ρ=
where 2 is a constant. ρ̂ is the estimation of b . D̂ is an estimation of D . D is the
unknown bound of d (t ) . The update laws can be chosen as


ρˆ
= −γ z2 v ; Dˆ = η | z2 |; aˆ
= µ z2 x2 (15)
where η , γ , µ are positive constants and can be designed arbitrarily. To analysis the stability of
closed loop system, the following Lyapunov function can be chosen
1 |b| 2 1 2 1 2
V = V1 + z22 + ρ + D + a
2 2γ 2η 2µ (16)
where D = D − D ,
 ˆ ρ = ρ − ρ
ˆ , a = a − aˆ are estimation errors of unknown constant parameters.
Note that the following inequality can be obtained easily
bv = v − b ρ v (17)
z2
With (13) and (14), the derivative of can be written as
z
2 = ax2 + bv + d (t ) − α
1 −


yr
= ax − z − ax
2 1
ˆ − k z + α
+


2 2 2 y − sgn( z ) Dˆ − b ρ v + d (t ) − α
−


1 r 2 y 1 r

= − z1 − k2 z2 + a x2 + d (t ) − sgn( z2 ) Dˆ − b ρ v
T
(18)
Then we can get
1
( z22 )′ = z2 z
2 = − z1 z2 − k2 z22 + a T z2 x2 + z2 d (t )− | z2 | Dˆ − b ρ vz2
2
≤ − z1 z2 − k2 z22 + a T z2 x2 + | z2 | D − b ρ vz2
(19)
From (12), (16), (19) the derivation of Lyapunov function can be rewritten as
 Applied Mechanics and Materials Vols. 88-89 91

| b |
1 
ˆ 1

V
= V
1 + z2 z
2 − ρρ
 ˆ − DD − aa
ˆ
γ η µ
2
| b |
1  ˆ
1

≤ −∑ ki zi2 + az
 2 x2 + | z2 | D − b ρ vz2 − ρρ
 ˆ − DD − aa
ˆ
i =1 γ η µ
2
1  ˆ
|b|
1
≤ −∑ ki zi2 − D( D − η | z2 |) − ρ ( ρˆ + sign(b)γ z2v ) − a (aˆ
− µ z2 x2 )
i =1 η γ µ (20)
With the update laws shown in (15), we can get the following inequality
2
V
≤ −∑ ki zi2
i =1 (21)
Theorem 1: Considering the system (1), with the control law (14) and the update laws (15) and the
Assumption 1-2. Then all signals in closed-loop system are bounded. The tracking error such
lim | y − yr |= 0
that t →∞ .
Prove: The Lyapunov function can be chosen as (16). From (21) the result of theorem 1 can be got
easily. Then according the LaSalle's invariant principle, the tracking performance can be obtained.

Simulation studies
In this section, we use the same model in [1]. The actual values of system unknown parameters
J = 0.067 , KT = 2.5 , K e = 0.25 , Ra = 0.4 , K = 200 and the known constant i = 1650 . The
backlash-like hysteresis is described by (4) . The parameters in backlash model are α = 1 ,
c = 3.1635 and B1 = 0.345 . So the unknown parameters in system model (7) can be calculated as
a = −23.32, b = 35.77 . The reference signal is sin(t ) . We chose k1 = k2 = 10 . In simulation the

initial value are chosen as follows: z (0) = [0.5, 0] , ρˆ (0) = 0, D (0) = 0, aˆ (0) = 0 , u (0) = 0 . The design
ˆ

parameters γ = η = µ = 0.1 . Fig.1 is tracking error, Fig.2 are actuator input v(t ) and output u (t )

Fig.1 Tracking error Fig.2 Backlash output and input

Conclusions
In this paper. an adaptive state feed back control scheme is developed for uncertain gun control
system of tank with backlash-like input. The stability of closed-loop system can be guaranteed by the
control scheme. But discontinuous function sign(⋅) is involved in the control laws which will cause
chattering. So our further research can focus on how to design continuous control law to avoid
chattering.
Acknowledgments: This work is supported by science foundation of Zhejiang Water Conservancy
And Hydropower College (xky-201102).
92 Computer-Aided Design, Manufacturing, Modeling and Simulation

References
[1] C. S. Sun, J. Chen, L. H. Zou,“Variable structure control for sliding mode of tank stabilizator
based on optimization”, Acta armamentarii, Vol. 22, 2001, p. 119.
[2] X. J. Ma, L. Feng, D. Yuan, et al.,“Overview of adaptive compensation control of nonlinearity in
tank gun control system”, Fire control and command control, Vol. 35, 2010, p. 1.
[3] F. Liang et al.,“Adaptive sliding mode robust control of gun control system of tank”, Fire control
and command control, Vol. 35, 2010, p. 55.
[4] C. Y. Su, Y. Stepanenko, J. Svoboda.,“Robust adaptive control of a class of nonlinear systems
with unknown backlash-like hysteresis”, IEEE Trans. Auto. Control, Vol. 45, 2000, p. 2427.
[5] J. Zhou, C. Wen, Y. Zhang,“Adaptive backstepping control of a class of uncertain nonlinear
systems with unknown backlash-like hysteresis”, IEEE Trans. Auto. Control, Vol. 49, 2004, p.
1751.
Computer-Aided Design, Manufacturing, Modeling and Simulation
10.4028/www.scientific.net/AMM.88-89

Adaptive Control of Uncertain Gun Control System of Tank

10.4028/www.scientific.net/AMM.88-89.88

DOI References
[4] C. Y. Su, Y. Stepanenko, J. Svoboda., Robust adaptive control of a class of nonlinear systems with
unknown backlash-like hysteresis, IEEE Trans. Auto. Control, Vol. 45, 2000, p.2427.
doi:10.1109/9.895588
[5] J. Zhou, C. Wen, Y. Zhang, Adaptive backstepping control of a class of uncertain nonlinear systems with
unknown backlash-like hysteresis, IEEE Trans. Auto. Control, Vol. 49, 2004, p.1751.
doi:10.1109/TAC.2004.835398