Author’s Accepted Manuscript

Application of active disturbance rejection control in

tank gun control system
Yuanqing Xia, Li Dai, Mengyin Fu, Chunming Li,
Chunming Wang

www.elsevier.com/locate/jfranklin

PII: S0016-0032(13)00060-4
DOI: http://dx.doi.org/10.1016/j.jfranklin.2013.02.003
Reference: FI1687

To appear in: Journal of the Franklin Institute

Received date: 9 April 2012

Revised date: 8 January 2013
Accepted date: 11 February 2013

Cite this article as: Yuanqing Xia, Li Dai, Mengyin Fu, Chunming Li and Chunming
Wang, Application of active disturbance rejection control in tank gun control system,
Journal of the Franklin Institute, http://dx.doi.org/10.1016/j.jfranklin.2013.02.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As a
service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting galley proof
before it is published in its final citable form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply
to the journal pertain.
 Application of Active Disturbance Rejection Control in
Tank Gun Control System

Yuanqing Xia, Li Dai, Mengyin Fu, Chunming Li and Chunming Wang

Abstract

The problem of position tracking for a tank gun control system with inertia uncertainty and external disturbance
is investigated in this paper. The tank gun control system, demanding high tracking precision and stabilization
precision, is a nonlinear system. Classical control methods are commonly used in the actual system, which is
difficult to ensure high precision and high disturbance rejection capability. An active disturbance rejection control
(ADRC) scheme is applied to guarantee the state variables of the closed loop system to converge to the reference
state with the help of the extended state observer by estimating the inertia uncertainty and external disturbance. The
basic theory of the ADRC is introduced here. According to the mathematical model, the parameters of ADRC are
designed. Also, simulation results show that ADRC controller has advantages of high precision and high disturbance
rejection ability. A comparison between ADRC and PID is also presented, to show the effectiveness of the ADRC
control strategy.

Index Terms

Tank gun, ADRC, SMC, uncertainties, disturbances

I. INTRODUCTION

Tank gun control system is an important means to improve the efficiency of extensive artillery firepower
and improve the surviving ability in the battlefield. An important feature of a military tank is the ability
to track the moving targets. The gun controller is a very important part of tank fire control system and
it has an important effect on its tactical and technical performance. The control of a tank turret and
gun is a formidable problem, because accuracy, stability and speed of response are essential to mission
accomplishment and survivability. Furthermore, the tank gun control system should not only have high
tracking precision and good dynamic quality, but also strong robustness to overcome the parameters

Yuanqing Xia, Li Dai, Mengyin Fu and Chunming Wang are with the School of Automation, Beijing Institute of Technology, Beijing
100081, China. Email: xia yuanqing@bit.edu.cn; daili1887@gmail.com; fumy@bit.edu.cn; chunming850313@gmail.com. Chunming Li is
with the China North Vehicle Research Institute, P. O. Box; 969 11, Beijing, China. E-mail:chunming@noveri.com.cn
perturbation and outside disturbance during operation. Indeed, some reasons can be found in the tank gun
control to confine the performance of this weapon, such as, moments of inertia and friction disturbance
torques change with the gun position, friction and other nonlinearities change over time, the disturbances
movement over rough terrain.
Tank gun is an important weapon and plays a vital role in wars. Much work has been done on the
design of tank gun control system, e.g. variable structure control [1], [2], optimal control [3], [4], fuzzy
control [5] and robust control [6]. [2] introduced sliding mode control combined with adaptive fuzzy
control of gun control system for tank tracking and disturbance rejection. The problem from a robotics
point of view and a minimum-time control law was given in [7]. [8] used an auto-regressive (AR) model
to model the gun fire control system and forecasted the future position of the tank based on velocity. [1]
showed variable structure control (VSC) with a sliding mode to solve robustness problem. [9] studied the
disturbance rejection problem for a linear model of the weapon using model predictive control due to its
ability to handle the constraints.
It is easy to describe the basic principle of tank gun stability, but it is also a very complex work to
achieve. Parameters perturbation, outside disturbance and many nonlinear factors exist in the regulator
system. In order to overcome these problems, a valid control scheme must be established to guarantee
system reliability and image quality. At present, each country employs different technique approaches to
improve the technique of the gun control system. Many kinds of modern control theories, to a great extent,
are not realistic because they demand an accurate mathematical model which is not possible to be obtained
in some specific conditions [10]. Recognizing the vulnerability of the reliance on accurate mathematical
model, there has been a gradual recognition over the years that active disturbance estimation is a viable
alternative to an accurate plant model. The focal point is how external disturbance and unknown dynamics
can be estimated. Several classes of approach are outlined below, including the unknown input observer
(UIO) [11], [12], the disturbance observer (DOB) [13], [14], the perturbation observer (POB) [15], and the
extended state observer (ESO) [16]-[20]. ESO can easy to estimate not only the external disturbance but
also plant dynamics that is different from UIO and DOB. Based on ESO, it is shown that, for an nth order
plant to be controlled, its internal dynamics and external disturbances can be estimated in real time from
its input-output data. It means that an accurate model of the plant is perhaps no longer required. Among
the aforementioned disturbance estimators, ESO requires the least amount of plant information[21].
ESO was first proposed in the context of the active disturbance rejection control (ADRC) [16]-[20],
which was firstly proposed by Han in [22] [23]. ADRC has been proposed and been in progress for
almost two decades, and its applications can be found in lots of literature in recent years. ADRC has
been introduced in different fields [24]-[27]. The application of ADRC and its actual implementation to
tank gun control is, we believe, novel. Active disturbance rejection controller (ADRC) is a new non-linear
algorithm used in servo system in recent years. It maintains the advantage of PID, and is introduced
into the motion control system by [28] and [29]. We suggest that there are four fundamental technical
limitations in the existing PID framework, and active disturbance rejection control or ADRC is very good
to deal with these problem, including the following: 1) a simple differential equation to be used as a
transient profile generator; 2) a noise-tolerant tracking differentiator; 3) the power of nonlinear control
feedback; and 4) the total disturbance estimation and rejection [20]. ADRC is also not predicated on an
accurate and detailed dynamic model of the plant and is extremely tolerant of uncertainties and simple
to use. By real-time estimates and compensation of system’s internal and external disturbance such as
the change of the load torque, combining with nonlinear control strategy, it can get better static and
dynamic performances, strong robustness and adaptability. [30] shows an example of a permanent magnet
synchronous motor position servo system based on ADRC. In [32], ADRC is introduced in pitch and roll
attitude control to solve some problems appeared in flight attitude control. In [33], a decoupling active
disturbance rejection control (ADRC) scheme for Integrated Flight-Propulsion Control (IFPC) is designed.
In [34], we discuss the problem that the helicopter manoeuvres naturally in the environment where the
execution of the task can be easily affected by atmospheric disturbances. In [35], we analyze the ship
course motion characteristics and give the nonlinear model with the disturbance and design an ADRC
controller. In [36], a ship main engine optimal ADRC controller is designed to solve the difficulty that
the mathematics model of nonlinear ship main engine controller, and the wave disturbances to the design
of electronic governor have the unmatched uncertainty correspondingly. In [37], a straight-line tracking
controller is designed for the nonlinear and under-actuated mathematical module of ships straight-line
tracking control system which is based on ADRC. In [38], the attitude control for a spacecraft model
which is nonlinear in dynamics with inertia uncertainty and external disturbances is discussed. Since the
ADRC does not depend on the accurate model of the servo system, it is very robust against parameter
variations, disturbances, and noises, not only in some operation areas but also in the whole working area.
The purpose of this paper is to show that ADRC control algorithm has advantages of high precision and
high disturbance rejection ability for tank gun control system with torque and disturbance uncertainty.
 The research work is focused on the servo system of tank gun and has a practical background. This
work tries to propose some efficient methods to solve the problems existing in the technique field. We will
test and verify the algorithm on the semi-physical simulation platform. The paper is organized as follows.
In Section II, the model of a tank gun control system is given, along with the mathematical model. Section
III shows the basic idea of ADRC strategy, and an extended state observer (ESO) is used to estimate the
plant dynamics and disturbances to actively compensate for the disturbance. The comparison of classical
PID and ADRC is given in Section IV, the characteristics and weaknesses of ADRC are also discussed,
and how this new framework is used to solve various kinds of control problems. Finally, some concluding
remarks are given in Section V.

II. PROBLEM STATEMENT

The block diagram of the gun control system, a careful reduction of a complex nonlinear simulation
model [31], is shown in Fig.1. This is a simplified servo system including speed loop and position loop.
An ADRC controller is employed here to track the reference input efficiently. It contains the following
variables and parameters:
La Armature inductance
Ra Armature resistance
Ia Armature current
Kt Torque coefficient
Ke Counter voltage factor
J Moment of inertia of the motor
JL Moment of inertia of the load
Te Comprehensive disturbance instead of all kinds of factors
Tem Motor output torque
θr reference position
θ position output
ωs angular velocity
The model for the tank gun control system is given by




 x1 = y


ẋ1 = x2 (1)




ẋ2 = f (x1 , x2 , ω(t), t) + bu



We define x1 = y = θ is output variable, x2 = ws , b is magnification factor and ω(t) is the external

disturbance. f (x1 , x2 , ω(t), t) is the total external and internal disturbances function.

Fig.1. Transfer function block diagram of servo system of tank gun

III. ADRC STRATEGY

The classical PID control theory focuses on eliminating the control error by use of the current (P), past
(I) and future (D) states of the feedback error.
Proportional-integral-derivative (PID) control law is:

Rt
u = K0 0 e(τ )dτ + K1 e + K2 ė

where e(τ )dτ, e and ė are integral of error, error and error change rate and belong to R n . K0 , K1 ,
Rt
0

K2 are integral gain, proportional gain and derivative gain coefficient and belong to R m×n . There is a
question, which means that if the load changes in a very large range, we can’t change the parameters
online to achieve the system’s request.
Generally, gun controllers are designed by classical control theory, such as PID, so the controllers
designed depend heavily on the models and when the plants parameters vary largely, its very difficult to
keep the original good dynamic and still performances.
It is well known that sliding mode control (SMC) is a robust method to control nonlinear and uncertain
systems which has attractive features to keep the system insensitive to the uncertainties on the sliding
surface. The control input u can be designed as follows:


 u+ (x) s(x) > 0
u=
u− (x) s(x) < 0



The sliding surface is shown in Fig. 2.

Fig.2. Attracted by sliding surface .

The conventional SMC design approach consists of two steps. First, a sliding manifold is designed such
that the system trajectory along the manifold acquires certain desired properties. Then, a discontinuous
control is designed such that the system trajectories reach the manifold in finite time. Sliding mode control
as a general design tool for control systems has been well established, the primary advantages of sliding
model control are: I) fast response and good transient performance; II) its robustness against a large class
of perturbations or model uncertainties; and III) the possibility of stabilizing some complex nonlinear
systems which are difficult to stabilize by continuous state feedback laws. In the last few years, so many
valuable results have been obtained on different aspects of SMC [39]-[46].
The ADRC originates from SMC, it is a novel innovation of SMC. However, the ADRC method which
doesn’t depend on system model can estimate and compensate the influences of all the internal and external
disturbances in real time. The ADRC has the same advantage of fast response and strong robustness as
traditional PID control theory, it gives a new control theory and control method, which is widely applied
for its excellent system performance. The ADRC consists of three parts, a nonlinear tracking differentiator
(TD), which is used to arrange the ideal transient process of the system, an extended state observer (ESO),
which could estimate all the disturbances from the system output, and then the ADRC compensates the
disturbance according to estimated values, a nonlinear state error feedback (NLSEF), which is used to
get the control input of the system. The structure of ADRC controller is shown in Fig.2. Consider system
(1), where x1 , x2 are state vectors, y is output variable, u is control variable. The ADRC approach makes
an effort to compensate for the unknown dynamics and external disturbances in the time domain.

Fig.3. The structure of ADRC Algorithm.

A. TD

The performance of some control system is restricted by the differential signal selected from the non-
continuous noisy measured signal. The differential signal is usually obtained by the backward difference
of the given signal, but, it will contain a certain amount of stochastic noise. However, TD has the ability to
resolve the problem of differential signal extraction via integration. Therefore, TD can avoid unnecessary
noise and make the system more effective and robust performance in some situations.
One feasible second-order TD can be designed as


 v̇1 = v2
v̇2 = f han(v1 − v(t), v2 , r, h0 )



where v(t) denotes the control objective. v1 is the desired trajectory and v2 is its derivative. r and h0
are controller’s parameters. r is speed factor and it will decide tracking speed. The greater value of r is,
the faster transition process will be. h0 is filtering factor, which makes an effort of filter. As we know,
decreasing the integration step will make an great influence on limiting the noise. When the integration
step is fixed, increasing the filtering factor will make the filter perform better. f han(v 1 − v(t), v2 , r, h0 ) is

d = rh20 , a0 = h0 v2 , y = (v1 − v(t)) + a0

q
a1 = d(d + 8|y|)
a2 = a0 + sign(y)(a1 − d)/2
sy = (sign(y + d) − sign(y − d))/2
a = (a0 + y − a2 )sy + a2
sa = (sign(a + d) − sign(a − d))/2
f han = −r( ad − sign(a))sa − rsign(a)

The TD is such a nonlinear component which provides transition process for expected input v and
differential trajectory of set value, that is v1 and its differential v2 . TD has the ability to track the given
input reference signal with quick response and no overshoot.

B. ESO

f (x1 , x2 , ω(t), t) generally includes three parts: modeling dynamics, uncertain dynamics (or uncertain
acceleration) and disturbance, it is difficult to get the exact model of f (x1 , x2 , ω(t), t) or its approximate.
The ESO is used to estimate f (x1 , x2 , ω(t), t) in real time and make adjustments at each sampling point
in a digital controller. Here f (x1 , x2 , ω(t), t) is considered as an extended state for the system, x3 as the
uncertainties f (x1 , x2 , ω(t), t), and its differential a(t).
We can rewrite this system as follows:

x1 = y







ẋ1 = x2




ẋ2 = x3 + bu








ẋ3 = a(t)



Then, we can use the following nonlinear observer to estimate x and f (x1 , x2 , ω(t), t).

e = z1 − y







ż1 = z2 − β1 e




ż2 = z3 − β2 f al(e, α1 , δ) + b0 u








ż3 = −β3 f al(e, α2 , δ)


where z1 , z2 and z3 are the observer outputs and β1 , β2 and β3 are observer gains. e is the error, z1 is used
to estimate system output, z2 is used to estimate the differential of system output, z3 is extended state
variable to estimate comprehensive disturbance. Parameters β1 , β2 and β3 must be tuned appropriately to
achieve the good performance.

C. Nonlinear Combination

State error feedback control law generates control voltage u for system by using the error from the
output of ESO and TD. The errors are combined with nonlinear manners, large errors correspond to lower
gains and small errors correspond to higher gains.


 e1 = x1 − z1
e2 = x2 − z2



A nonlinear combination of error signal and its differential can be constructed as follows:

u0 = k1 f al(e1 , α1 , δ) + k2 f al(e2 , α2 , δ)

where k1 and k2 are proportional and differential coefficients respectively, f al(x, α, δ)is a nonlinear
function described by 
x


δ1−α
, |x| ≤δ
f al(x, α, δ) =
sign(x)|x|α , |x| > δ



The nonlinear function is used to make the observer more efficient. In order to achieve better performance,
the nonlinear coefficient α1 and α2 are selected as 0 < α1 < 1 < α2 .
The controller is designed as
z3
u = u0 − .
b0

IV. E XPERIMENTAL R ESULTS

Here we consider the following second-order system.



 ẋ(t) = Ax(t) + Bu(t) + Df (t)
(2)
y(t) = Cx(t)


where
     
0 1 0 h i 0
A= , B= , C= 1 0 , D= ,
a1 a2 b d
a1 = −(Ke ∗ Kt )/La (J + JL ), a2 = −Ra /La , b = 1/(La (J + JL ) ∗ Ke ).

f (t) ∈ R denotes the external disturbance and the coefficient d ∈ R denotes the system uncertainty.

Ra = 0.279Ω, La = 0.66mH,

Kt = 0.179NM/A, Ke = 0.179V rad/s,

J = 1.378 × 10−3 NM 2

In order to control the servo system and achieve the objective, the speed control and the position control
are considered respectively. In this paper the ADRC and the conventional PID control are studied and
simulated aimed at (1) in order to illustrate the feasibility and effectiveness of the ADRC technique. Also
the results of the proposed controller is presented in this section, which can be seen in Figs. 3,4,5,6,7 and
8. The parameters of the system are listed as follows: b1 = 0.5, b2 = 0.5, b3 = 0.5, b01 = 0.02, b02 = 0.005.
Because of the parameter robustness, it is suitable for the following system.

A. The speed control

Here we will just focus on the inner loop, which is also the speed loop. Assuming the closed loop system
without disturbance (Te ) and without torque(JL ). With reference input, actual speed changes are based on
Fig. 1, where the different control strategy with ADRC and PID are shown respectively. Assuming the
input equal 1 and the disturbance and torque are discussed respectively.
Fig.3 shows the response of the system without disturbance and without torque and the closed loop
transfer function can be written as case 1; Fig.4 shows the response of the system without disturbance
and with torque and the closed-loop transfer function can be written as case 2; Fig.5 shows the response
of the system with disturbance and without torque and the closed-loop transfer function can be written
as case 3; Fig.6 shows the response of the system with disturbance and with torque and the closed-loop
transfer function can be written as case 4.
When the disturbance is ignored, the simulation result shows that in speed control, the response of the
ADRC is similar to that of PID controller when adding the moment of inertia of the load. The simulation
result also shows the same when disturbance and moment of inertia of the load are ignored together.
When adding the disturbance, the curve shows differently. The ADRC has a small overshooting on the
controlling, while, the PID control effect becomes very bad. It demonstrates that the ADRC has obvious
advantages compared with PID control on disturbance rejection in control system.
case 1 JL = 0, Te = 0
5.59
G(s) =
0.0000288s2 + 0.012s + 1

case 2 JL = 0.001sin(t), Te = 0

5.59
G(s) =
(0.0000288 + 0.02592JL )s2 + c1 + 1

with
c1 = (0.012 + 8.7048JL )s

case 3




 30000, t = k + 0.001s


JL = 0, Te = −30000, t = k + 0.002s




0, else



5.59(1 + 0.001378Te)
G(s) =
0.0000288s2 + 0.012s + 1 + 0.001378Te

case 4




 30000, t = k + 0.001s


JL = 0.001sin(t), Te =  −30000, t = k + 0.002s
 

0, else



5.59[1 + (0.001378 + JL )Te ]

G(s) =
(0.0000288 + 0.02592JL )s2 + c2 + 1 + c3

with
c2 = (0.012 + 8.7048JL )s, c3 = (0.001378 + JL )Te
 1.5

ADRC
0.5

0
0 1 2 3 4 5 6 7 8 9 10
Time(s)

1.5

PID

0.5

0
0 1 2 3 4 5 6 7 8 9 10
Time(s)

Fig.4. system response without disturbance and without torque.

1.5

1
ADRC

0.5

0
0 1 2 3 4 5 6 7 8 9 10
Time(s)

1.5

1
PID

0.5

0
0 1 2 3 4 5 6 7 8 9 10
Time(s)

Fig.5. system response without disturbance and with torque.

1.5

1
ADRC

0.5

0
0 1 2 3 4 5 6 7 8 9 10
Time(s)

1.5

1
PID

0.5

0
0 1 2 3 4 5 6 7 8 9 10
Time(s)
 Fig.6. system response with disturbance and without torque.

1.5

ADRC
0.5

0
0 1 2 3 4 5 6 7 8 9 10
Time(s)

1.5

1
PID

0.5

0
0 1 2 3 4 5 6 7 8 9 10
Time(s)

Fig.7. system response with disturbance and with torque.

B. The position control

The position control is a double closed-loop control including speed loop and position loop. Assuming
the input equal 1 and the system with disturbance and the moment of inertia of the load. Firstly, the
Proportional control is given in position loop, and the closed-loop transfer function can be written as case
5. The speed loop with ADRC and PID control will be discussed respectively, which is shown in Fig.7.
We can see that the PID and ADRC show also the same performance under the same condition. Secondly,
the Proportional control is given in speed loop, and the closed-loop transfer function can be written as
case 6. The position loop with ADRC and PID control will be discussed respectively, which is shown in
Fig.8. It can be clearly seen that the system with ADRC control in position loop behaves smoothly on the
real speed, while the real speed curve by PID controller fluctuates suddenly. This is mainly because the
ADRC technology uses ESO to observe the disturbances accurately and makes accurate compensation to
improve the precision and response speed, which don’t need the integrator to eliminate the steady-state
error. Both the position loop and the speed loop with ADRC are shown in Fig.9.
case 5




 7000, t = k + 0.001s


JL = 0.001sin(t), Te = −7000, t = k + 0.002s




0, else


 5.59[1 + (0.001378 + JL )Te ]
G(s) =
(0.0000288 + 0.02592JL)s2 + c3 + c4

with
c3 = (0.012 + 8.7048JL )s, c4 = 1 + (0.001378 + JL )Te

case 6




 38000, t = k + 0.001s


JL = 0.0001sin(t), Te = −38000, t = k + 0.002s




0, else



5.59[1 + (0.001378 + JL )Te ]

G(s) =
(0.0000288 + 0.02592JL)s2 + c5 + c6

with
c5 = (0.012 + 8.7048JL )s, c6 = 1 + (0.001378 + JL )Te

2
reference position
real position
reference speed
1.5 real speed
ADRC

0.5

0
0 1 2 3 4 5 6 7 8 9 10
time(s)

2
reference position
real position
reference speed
1.5 real speed
PID

0.5

0
0 1 2 3 4 5 6 7 8 9 10
time(s)

Fig.8. The position loop with P control.

2
reference position
real position
real speed
1.5
ADRC

0.5

0
0 1 2 3 4 5 6 7 8 9 10
time(s)
 2
reference position
real position
real speed
1.5

PID
1

0.5

0
0 1 2 3 4 5 6 7 8 9 10
time(s)

Fig.9. The speed loop with P control.

2
reference position
real position
real speed
1.5
ADRC

0.5

0
0 1 2 3 4 5 6 7 8 9 10
time(s)

Fig.10. Both the speed loop and the position loop with ADRC control.

V. CONCLUSIONS AND FURTHER REMARKS

The paper has adopted ADRC algorithm for tank gun with torque and disturbance uncertainty. It
has been shown that ADRC control strategy is independent of system model and external disturbance.
The performances of ADRC and conventional PID controller are compared under the same situation.
Simulation results of ADRC control system and PID control system show that the ADRC controller has
good dynamic and static characteristics, while the classical PID control is hard to guarantee high precision
and high disturbance rejection ability with the existence of disturbance. Using ESO in the feedback path to
estimate the unmeasured state and the real action of the unknown disturbances and compensated exactly in
real time. This indicates that the ADRC algorithm can overcome the disturbance, having a good tracking
and robust performance, which provides an effective method for servo system control. It is believed that
the application of ADRC in the tank gun will surely be further acknowledged, along with the development
of the ADRC theory.
In a real system, unknown actuator/sensor fault or time delay in system states always exist. Hence,
simultaneous state and fault estimation is more desirable. In [47], the problem of observer-based controller
design against sensor faults for nonlinear Markovian jump system with time delay is investigated. There
may be some difficulties for combing the techniques in this paper with [47]. Further, in [48], the authors
have proposed a new observer approach to obtain the simultaneous estimates of system states, sensor fault
and actuator fault vectors, and an observer-based fault-tolerant control scheme is developed to stabilize
the resulting closed-loop system. These techniques may be incorporated for studying ADRC in tank gun.

ACKNOWLEDGEMENTS

The authors would like to thank the referees for their valuable and helpful comments which have
improved the presentation. The work of Yuanqing Xia was supported by the National Basic Research
Program of China (973 Program) (2012CB720000), the National Natural Science Foundation of China
(61225015,61127004,60974011), Program for New Century Excellent Talents in University of China
(NCET-11-0784), the Ph.D. Programs Foundation of Ministry of Education of China (20091101110023,
20111101110012), Beijing Municipal Natural Science Foundation (4102053,4101001), and CAST Foun-
dation (CAST201210).

R EFERENCES

[1] R. Dana and E. Kreindler, “Variable Structure Control of a Tank Gun”, Proceedings of Control Applications, Dayton, OH, 1992,
pp.928-933.
[2] L. Feng, X. J. Ma, Z. F. Yan and H. Li, “Method of adaptive fuzzy sliding mode control of gun control system of tank”, Dianji yu
Kongzhi Xuebao/Electric Machines and Control, vol.11, no.1, 2007, pp.65-69.
[3] W. Grega, “Time-optimal control of n-tank system”, Proceedings of the 1998 IEEE International Conference on Control Applications,
Trieste, 1998, pp.522-526.
[4] S. Song, G. Cai, B. Liao and X. Lin, “Real-time optimal control for three-tank level system via improved ADDHP method”, Proceedings
of Control and Automation, Christchurch, 2009, pp.564-568.
[5] H. Lin, Z. Teng, T. Chen, S. Yang, H. Zhang and H. Chi, “Improved fuzzy control method for temperature in water tank of intelligent
viscometer”, Proceedings of Information and Automation, Changsha, China, 2008, pp.1106-1110.
[6] W. Chai, N. K. Loh, C. F. Lin and N. Coleman, “Robust Digital Control of Gun Turret Systems”, Proceedings of American Control
Conference, Chicago, IL, USA, 1992, pp.419-423.
[7] F. L. Lewis, D. M. Dawson, J. Lin and K. Liu, “Tank gun-pointing control with barrel flexibility effects”, Proceedings of Winter Annual
Meeting of the American Society of Mechanical Engineers, ASME Dynamic System and Control division, Atlanta, GA, 1991, pp.65-69.
[8] S. G. Kapoor, C. Kunpanitchakit, S. M. Wu and N. Coleman, “Modeling and forecasting of maneuvering tank position for gun fire
control”, in Modeling and Simulation, Proceedings of the Annual Pittsburgh Conference, 1979, pp.197.
[9] G. Kumar, P. Y. Tiwari, V. Marcopoli and M. V. Kothare, “A Study of a Gun-Turret Assembly in an Armored Tank using Model
Predictive Control”, Proceedings of American Control Conference, St. Louis, MO, 2009, pp. 4848-4853.
[10] Z. Gao and R. R. Rhinehart, “Theory vs. practice forum”, Proceedings of American Control Conference, Boston, MA, 2004, pp.1341-
1349.
[11] G. Basile and G. Marro, “On the Observability of Linear, Time-Invariant Systems With Unknown Inputs”, Journal of Optimization
Theory and Applications, vol.3, no.6, 1969, pp.410-415.
[12] J. Chen, R. J. Patton and H. Zhang, “Design of Unknown Input Observers and Robust Fault Detection Filters”, International Journal
of Control, vol.63, no.1, 1996, pp.85-105.
[13] R. Bickel and M. Tomizuka, “Passivity-Based Versus Disturbance Observer Based Robot Control: Equivalence and Stability”, Journal
of dynamic systems, measurement, and control , vol.121, no.1, 1999, pp.41-47.
[14] E. Schrijver and J. Dijk, “Disturbance Observers for Rigid Mechanical Systems: Equivalence, Stability, and Design”, Journal of Dynamic
Systems, Measurement, and Control , vol.124, no.4, 2002, pp.539-548.
[15] S. Kwon and W. K. Chung, “Combined Synthesis of State Estimator and Perturbation Observer”, Journal of dynamic systems,
measurement, and control, vol.125, no.1, 2003, pp.19-26.
[16] J. Han, “A Class of Extended State Observers for Uncertain Systems”, Control and Decision, vol.10, no.1, 1995, pp.85-88.
[17] Z. Gao, Y. Huang and J. Han, “An Alternative Paradigm for Control System Design”, Proceedings of IEEE Conference on Decision
and Control, Orlando, FL, 2001, pp.4578-4585.
[18] Z. Gao, “Scaling and Bandwidth-Parameterization Based Controller Tuning”, Proceedings of the American Control Conference, 2003,
pp.4989-4996.
[19] Z. Gao, “Active Disturbance Rejection Control: A Paradigm Shift in Feedback Control System Design”, Proceedings of the American
Control Conference, Minneapolis, MN, 2006, pp.2399-2405.
[20] J. Han, “From PID to Active Disturbance Rejection Control”, IEEE Transactions on Industrial Electronics, vol.56, no.3, 2009, pp.1-7.
[21] Q. Zheng, Q. Gao and Z. Gao, “On Validation of Extended State Observer Through Analysis and Experimentation”, Journal of Dynamic
Systems, Measurement, and Control, vol.134, no.2, 2012, pp.024505 (6 pages) .
[22] J. Han, “Active disturbance rejection control and its application”, Control and Decision, vol.13, no.1, 1998, pp.19-23.
[23] J. Han, “Nonlinear design methods for control systems”, Proceedings of the 14th IFAC World Congress, Beijing, China, 1999, pp.521-
526.
[24] Y. X. Su, B. Y. Duan, C. H. Zheng, Y. F. Zhang, G. D. Chen and J. W. Mi, “Disturbance-rejection high-precision motion control of a
stewart platform”, IEEE Transactions on Control System Technology, vol.12, no.3, 2004, pp.364-374.
[25] J. Su and W. Qiu,“Calibration-free robotic eye-hand coordination based on an auto disturbance rejection controller”, IEEE Transactions
on Robotics and Automation, vol.20, no.5, 2004, pp.899-907.
[26] Z. Ping and Z. Gao, “A FPGA-based digital control and communication module for space power management and distribution systems”,
Proceedings of the 2005 American Control Conference, 2006, pp.4941-4945.
[27] Y. Huang, Z. W. Luo and M. Svinin, “Extended state observer based technique for control of robot systems”, Preseeding of the 4th
World congress on Intelligent Control and Automation, Shanghai, China, 2002, pp.2807-2811.
[28] G. Feng, L. P. Huang and D. Q. Zhu, “High performance control of induction motor based on auto-disturbance rejection controller”,
Proceedings of the CSEE, vol.21, no.10, 2001, pp.55-58.
[29] Z. Q. Gao, S. H. Hu and F. J. Jiang, “A novel motion control design approach based on active disturbance rejection”, Proceedings of
IEEE Conference on Decision and Control, Orlando, FL, 2001, pp.4877-4882.
[30] K. Sun, Z. Xu, J. Zou and R. Dou, “A novel position controller of PMSM servo system based on active disturbance rejection controller”,
Proceeding of the CSEE, vol.27, no.15, 2007, pp.43-46.
[31] R. Dana, “Nonlinear Robust (VSC Based) Position Control of a Tank Gun”, Thesis (in Hebrew), Technion - Israel Institute of Technology,
March, 1991.
[32] M. Sun, Z. Chen and Z. Yuan, “A Practical Solution to Some Problems in Flight Control”, Proceedings of Joint 48th IEEE Conference
on Decision and Control and 28th Chinese Control Conference, Shanghai, China, 2009, pp.1482-1487.
[33] J. Wang, L. He and M. Sun, “Application of Active Disturbance Rejection Control to Integrated Flight-Propulsion Control”, Proceedings
of Chinese Control and Decision Conference, Xuzhou, China, 2010, pp.2565-2569.
[34] A. Martini, F. Leonard and G. Abba, “Robust nonlinear control and stability analysis of a 7DOF model-scale helicopter under vertical
 wind gust”, Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems Acropolis Convention Center, Nice,
France, 2008, pp.354-359.
[35] J. Ruan, Z. Li and F. Zhou, “ADRC Based Ship Course Controller Design and Simulations”, Proceedings of 2007 IEEE International
Conference on Automation and Logistics, Jinan, China, 2007, pp.2731-2735.
[36] W. Pan, Y. Zhou and Y. Han, “Design of Ship Main Engine Speed Controller Based on Optimal Active Disturbance Rejection Technique”,
Proceedings of the 2010 IEEE International Conference on Automation and Logistics, Hong Kong and Macau, 2010, pp.528-532.
[37] H. Xiao, C. Wang, Y. Han and W. Pan, “Design of Ship Straight-line Tracking Controller Based on Auto Disturbance Rejection
Control Technique”, Proceedings of 2010 8th World Congress on Intelligent Control and Automation (WCICA), Jinan, China, 2010,
pp.2559-2564.
[38] Z. Zhu, Y. Xia and M. Fu, “Attitude Tracking of Rigid Spacecraft Based on Extended State Observer”, Proceedings of 2010 3rd
International Symposium on Systems and Control in Aeronautics and Astronautics (ISSCAA), Harbin, China, 2010, pp.621-626.
[39] V. I. Utkin, “Variable structure systems with sliding modes”, IEEE Transactions on Automatic Control, vol.22, no.2, 2001, pp.212-222.
[40] W. Gao, Y. Wang and A. Homaifa, “Discrete-time variable structure control systems”, IEEE Transactions on Industrial Electronics,
vol.42, no.2, 1995, pp.117-122.
[41] M. V. Basin, A. G. Loukianov and R. Hernndez-Fabin, “An optimal sliding mode-like regulator for nonlinear polynomial systems”,
International Journal of Systems Science, vol.42, no.11, 2011, pp.1909-1916.
[42] K. Lu, Y. Xia, Z. Zhu and M. V. Basin, “Sliding mode attitude tracking of rigid spacecraft with disturbances”, Journal of the Franklin
Institute, vol.349, no.2, 2012, pp.413-440.
[43] L. Wu, X. Su and P. Shi, “Sliding Mode Control with Bounded L2 Gain Performance of Markovian Jump Singular Time-Delay
Systems”, Automatica, vol.48, no.8, 2012, pp.1929-1933.
[44] L. Wu and D. W. C. Ho, “Sliding Mode Control of Singular Stochastic Hybrid Systems”, Automatica, vol.46, no.4, 2010, pp.779-783.
[45] L. Wu, P. Shi and H. Gao, “State Estimation and Sliding Mode Control of Markovian Jump Singular Systems” , IEEE Transactions
on Automatic Control, vol.55, no.5, 2010, pp.1213-1219.
[46] X. Hua, L. Wu, C. Hu and H. Gao, “Adaptive Sliding Mode Tracking Control for a Flexible Air-breathing Hypersonic Vehicle”, Journal
of The Franklin Institute, vol.349, no.2, 2012, pp.559-577.
[47] M. Liu, P. Shi, L. Zhang and X. Zhao, “Fault tolerant control for nonlinear Markovian jump systems via proportional and derivative
sliding mode observer technique”, IEEE Transactions on Circuits and Systems, vol.58, no.11, 2011, pp.2755-2764.
[48] M. Liu, X. Cao and P. Shi, “Fuzzy-model-based fault tolerant design for nonlinear stochastic systems against simultaneous sensor and
actuator faults”, IEEE Transactions on Fuzzy Systems, (DOI: 10.1109/TFUZZ.2012.2224872).