Journal of Electrical Engineering & Technology

https://doi.org/10.1007/s42835-019-00256-0

ORIGINAL ARTICLE

Disturbance Observer Based Robust Sliding Mode Control

of Permanent Magnet Synchronous Motor
Nihad Ali1 · Ateeq Ur Rehman1 · Waqar Alam1 · Hamid Maqsood2

Received: 3 February 2019 / Revised: 25 May 2019 / Accepted: 2 August 2019

© The Korean Institute of Electrical Engineers 2019

Abstract
This paper addresses speed control of permanent magnet synchronous motor under load torque perturbations. The mathemati-
cal model is derived using park’s transformation. The load torque disturbance is considered unknown bounded, and states
variables are available in feedback. In order to achieve robust speed performance, sliding mode control (SMC) is introduced.
However, it is noted that conventional SMC does not provide satisfactory performance under load torque disturbances. To
end this, a novel control strategy called disturbance observer SMC (DOSMC) is formulated. It includes an observer that
offers a tool to vanish the effect of load torque. The DOSMC technique has two distinguished characteristics: first, the design
gains are needed to be greater than the maximum limit of disturbance estimation error instead of disturbances, second; the
proposed observer estimates load disturbances and provide a compensator to update sliding surface and control input. The
stability analysis of overall control system is verified using Lyapunov theorem. Simulations in MATLAB/Simulink proves
efficacy of the proposed scheme.

Keywords Disturbance observer · PMSM · Lyapunov · Robust · SMC

1 Introduction of the plant, and may not meet high performance require-
ment of speed tracking [4].
Permanent magnet synchronous motors (PMSM) have Therefore, modern nonlinear control strategies have been
significant advantages, that attracts the interest of many developed to achieve desire performance of industrial drive
researchers. It has wide applications in industry due to high application. Fuzzy logic-based controllers show robustness
efficiency, large torque-to-inertia ratio and superior power to parametric uncertainties as the design is independent of
density [1, 2]. Conventional linear controllers e.g. propor- the system model [5]. In this framework, expert knowledge
tional–integral–derivative and pole placement are commonly is essential for real-time application of such controllers
used in many applications [3]. However, these controllers are which is an impediment in implementation. In [6], artificial
sensitive to external disturbances, parametric perturbations neural network was presented based on adaptive controller
for the speed control of a PMSM. The results were conclu-
sive; conversely the control algorithm is complex. The adap-
* Nihad Ali tive backstepping is a recursive design technique [7, 8]. The
nihad.ali01@gmail.com use of virtual control variables is the most appreciating idea
Ateeq Ur Rehman that makes the high order system simple [9]. In addition, the
ateeq720@gmail.com adaptive laws eradicate the problem of parametric uncertain-
Waqar Alam ties. However, the mathematical derivation of controller and
engrwaqar423@gmail.com tuning the gains of all the estimator is tedious job. Feedback
Hamid Maqsood linearization is another popular method among control com-
hamid_maqsood@mail.nwpu.edu.cn munity and have been applied for the industrial machine
drives [10, 11]. In this method, some useful non-linearities
1
Department of Electrical Engineering, COMSATS may cancel out.
University, Islamabad 44000, Pakistan
Sliding mode control (SMC) has been widely used due
2
School of Automation, Northwestern Polytechnical to its strong disturbance rejection capability [12, 13]. The
University, Xi’an, China

13
Vol.:(0123456789)
 Journal of Electrical Engineering & Technology

trajectories motion on the sliding surface is not effected by dis-

R v
turbances that comes though input channel so-called matched i̇ d = − id − iq p𝜔 + d (3)
L L
disturbances [14]. However, load torque disturbances in
PMSM do not satisfy matching condition and performance is The dynamical model of PMSM is highly nonlinear. The
severely affected by such uncertainties [15–17]. Integral-SMC load torque disturbances are unknown and vary with time
is proclaimed to be robust and simplified controller [18]. How- depending on operating conditions. The various symbols and
ever, integral term in sliding surface causes some unfavora- nomenclature used in this paper are provided in Table 1.
ble effects such as long settling time and overshoot. In linear
matrix inequality based sliding surface design, the mismatched Assumption 1 The uncertain parameter 𝜏 in (1) is bounded
disturbances are assumed to be vanishing [19]. However, such and is given by |𝜏| ≤ 𝜏 ∗ where 𝜏 ∗ is the maximum limit of
assumptions are not realizable for real systems. Disturbance load torque disturbances.
observer-based state feedback controller [20] is suggested
for a linearized system. The controller shows robustness to To achieve high performance, the unknown load torque
disturbances. However, Jacobian [21] linearizing approach disturbances and all the nonlinearities must be considered in
may eradicate useful nonlinearities. In addition, the controller the controller design.
is sensitive to mismatch uncertainties. In [22], an observer
based on Lyapunov function is designed to estimate the load 3 Sliding Mode Control
torque. The disturbances under consideration are assumed to
be constant unknown. However, perturbations in load with SMC is a nonlinear robust control methodology. It is used
time might affect speed tracking performance. Controlling to force the system trajectories at the sliding manifold in the
speed of PMSM in the presence of load torque disturbances is presence of bounded uncertainties/disturbances. Control tech-
a challenging problem. niques based on SMC evolved from variable structure system
Motivated by abovementioned discussion, this paper pre- theory has attracted many researchers and designers due to its
sents a novel DOSMC design. The aim is to mitigate the effect robust nature against the system parametric uncertainties and
of load torque perturbations and control the speed of PMSM. external disturbances. Consider the general sliding manifold as
The disturbances under consideration are non-vanishing and
do not fulfil H2 norm bounded condition. First, a traditional
s1 = c1 z1 + ż 1 (4)
SMC is designed for speed tracking of PMSM and the short- where z1 = 𝜔 − 𝜔d is tracking error between actual and
comings of this technique are discussed. Then, by designing desire angular speed and c1 > 0 is arbitrary variable. Using
disturbance observer speed error trajectory can be driven (1) surface can be written as:
to origin asymptotically in the presence of mismatch load 3p𝜆f
uncertainty. To drive the initial trajectories on sliding surface s1 = c1 z1 + iq − b𝜔 − 𝜔̇ d (5)
2J
a discontinuous control law is designed. Robustness and quick
nominal performance recovery are the promising features of While substituting (1) the unknown term 𝜏 is omitted to
this control methodology. make the surface practically realizable. We need to drive (5) to
This paper is constituted of the following sections. Sec- zero in finite time by using control input vq for the convergence
tion 2 introduces the mathematical model of PMSM. In
Sects. 3 and 4 the SMC and proposed DOSMC controllers are
designed and performance analysis are provided for PMSM. Table 1  Nomenclature of parameters and states of PMSM
Simulation results are presented in Sect. 5, and finally conclud- Symbol Description Value Unit
ing remarks are given in Sect. 6.
𝜔 Rotor angular speed – Rad/s
id d-axis current – A
2 Mathematical Model Iq d-axis current – A
P Number of poles 2 –
Using parks transformation, the mathematical model of R Resistance 2.875 Ω
PMSM can be describe in dq [23] frame as L Inductance 0.0085 H
3p𝜆f 𝜆f Rotor to stator flux linkage 0.75 N m/A
𝜔̇ = iq − b𝜔 − 𝜏 (1) b Damping to inertia ratio 0.125 N s/kg/m
2J
𝜏 Load torque to inertia ratio – N/kg/m
𝜆f p vq B Damping constant 0.0001 Nms
R
i̇ q = − iq − id p𝜔 − 𝜔+ (2) J Inertia 0.0008 kg m2
L L L
TL Load torque – Nm

13
Journal of Electrical Engineering & Technology

of tracking error and error dynamic to zero in the presence second condition, we compute the time derivative of (13)
of bounded disturbance. This can be achieved by taking time along (8) and (12) as
derivative of surface using (1) and (2), which yields ( )
( ) V̇ = − k1 |s1 | − k2 s21 − c1 − b 𝜏s1 − k3 |s2 | − k4 s22
3p𝜆f R p𝜆f vq { ( ) } (14)
ṡ 1 = − iq − pid 𝜔 − 𝜔+ − 𝜔̈ d ≥ − |s1 | k1 + c1 − b 𝜏 ∗ − k2 s21 − k3 |s2 | − k4 s22
2J L L L ( )
( ) (6) By selecting the control gains k1 > c1 − b 𝜏 ∗ and
( ) 3p𝜆f
− c1 𝜔̇ d + c1 − b i − b𝜔 − 𝜏 k2 , k3 , k4 > 0 that makes V̇ < 0. Taking condition s1 = 0, at
2J q no load torque the solution of (4) can be computed as
z1 (t) = z1 (0)e−c1 t (15)
The speed and q-axis current trajectories can be stabilized
on sliding surface by designing the control input as The tracking error will reach the sliding surface in a finite
time and keep the tracking error at zero thereafter. If at certain
{ ( )
2JL ( ) 3p𝜆f time some load is attached with the motor shaft then, the solu-
vq = b − c1 i − b𝜔 tion of surface becomes as
3p𝜆f 2J q
( ) } (7) 𝜏 ( −c1 t )
+ c1 𝜔̇ d + 𝜔̈ d − k1 sgn s1 − k2 s1 z1 (t) = e − 1 + z1 (0)e−c1 t
c1 (16)
+ Riq + p𝜆f 𝜔 + Lpid 𝜔

Remarks The expression (16) shows that tracking error

where k1 and k2 are control gains. Back-substituting (7) into cannot be driven to surface even the control input is try-
(6), gives ing to converge the tracking error to equilibrium point.
( ) ( ) This is because the SMC surface manifold design cannot
ṡ 1 = − k1 sgn s1 − k2 s1 − c1 − b 𝜏 (8)
compensate mismatched disturbances. In other words, con-
The secondary manifold defined in (9), is considered for
ventional SMC is sensitive to load torque perturbation. The
q-axis current
actual speed deviates from its desire trajectory and causing
s2 = c2 z2 (9) a steady state error. The effect becomes more severe with
where c2 > 0 is an arbitrary variable and z2 = id − id_d is further increase in such disturbances. Moreover, the gain
d-axis current error with desire trajectory id_d = 0. We can requirement is high to achieve stability, which results in
stabilize secondary surface by designing control input vd . unwanted chattering phenomenon. To overcome such draw-
This can be accomplished by taking time derivative of (9) backs, DOSMC is proposed in the next section.
and using (3)
( 4 Disturbance Observer SMC
R v )
ṡ 2 = c2 − id + iq p𝜔 + d (10)
L L In this section a novel DOSMC control technique is proposed
The control law expression can be chosen as to mitigate is the effect of load disturbances. It combines the
conventional SMC and an internal state observer law that esti-
L{ }
mates instantaneous value of the disturbance and update the
vd = Rid − Lpiq 𝜔 − k3 sgn(s2 ) − k4 s2 (11)
c2 parameter in the control input. Surface manifold variable for
Substituting the control law (11) in (10), gives DOSMC can be defined as
( ) 3p𝜆f
ṡ 2 = − k3 sgn s2 − k4 s2 (12) s1 = c1 z1 + i − b𝜔 − 𝜏̂ − 𝜔̇ d (17)
2J q
3.1 Stability Analysis where 𝜏̂ is estimated disturbance. We need to drive speed
tracking error to defined surface in finite time under uncer-
In order to prove the stability of dynamics of (8) and (12) tain load torque by designing a control input vq. This can be
around and s1 = 0 and s2 = 0. A positive definite Lyapunov accomplished by taking time derivative of (17) along (1) and
function V can be considered as (2) one may have
1 2 1 2 ( )
V= s + s (13) 3p𝜆f R p𝜆f vq
2 1 2 2 ṡ 1 = − iq − pid 𝜔 − 𝜔+
2J L L L
which must satisfy the conditions i.e. V > 0 and V̇ < 0 for (
(
) 3p𝜆f
)
(18)
s1 , s2 ≠ 0, the first condition is obvious. To validate the + c1 − b iq − b𝜔 − 𝜏 − 𝜏̂̇ − 𝜔̈ d
2J
− c1 𝜔̇ d

13
 Journal of Electrical Engineering & Technology

We can stabilize surface and surface dynamics by design- It can be observed from (26) that if the observer tuning
ing the control law as gain satisfies the condition l < 0, the estimated load torque
{ ( ) can track actual load torque disturbances i.e. 𝜏̂ → 𝜏 or in
2JL ( ) 3p𝜆f
vq = b − c1 i − b𝜔 − 𝜏̂ other words, estimator error decays exponentially to origin.
3p𝜆f 2J q
( ) } (19)
+ c1 𝜔̇ d + 𝜔̈ d − k1 sgn s1 − k2 s1 4.1 Stability Analysis
+ Riq + p𝜆f 𝜔 + Lpid 𝜔
The Lyapunov function for DOSMC is same as (13) and
its time derivative using (20), (26) and (12) followed by
Substituting the control law (19) into (18), the surface
Assumption 3, can be expressed as
dynamics becomes
( ) ( ) ( )
ṡ 1 = − 𝜏̂̇ − k1 sgn s1 − k2 s1 − c1 − b 𝜏̃ (20) V̇ = s1 𝜏̃ l − c1 + b − k1 |s1 | − k2 s21 − k3 |s2 | − k4 s24
{ ( ) }
Similarly, the secondary surface and the d-axis voltage ≥ − |s1 | −𝜏̃ ∗ l − c1 + b + k1 − k2 s21 − k3 |s2 | − k4 s24
are same as (9) and (11). To achieve nominal performance (27)
of the control system an estimator that updates load torque The tracking error will reach zero in finite time
in (17) and (19), is introduced and depicted in [20] and remains
( thereafter
) as long as the designed gains
k1 > l − c1 + b 𝜏̃ ∗ and k2 , k3 , k4 > 0. At no load torque the
𝜏̂ = Q + l𝜔 (21) solution of (17), taking condition s1 = 0
where Q and l are internal state of disturbance observer and
tuning gain to be designed respectively. Define estimator z1 (t) = z1 (0)e−c1 t (28)
error The speed tracking error will reach the sliding surface and
remain at zero thereafter. If at certain time load is couples
𝜏̃ = 𝜏 − 𝜏̂ (22) with the shaft of motor the solution of surface becomes
The following assumptions are supposed to satisfy for
disturbance estimation. 𝜏̃ ( −c1 t )
z1 (t) = e − 1 + z1 (0)e−c1 t (29)
c1
Assumption 2 The time derivative of load disturbances is
bounded and holds the condition limt→∞ 𝜏̇ = 0. Remarks The expression (29) shows that speed tracking
error can be driven to origin. Since the disturbance has
Assumption 3 The disturbance estimator error in (22) is also been updated in control law by the designed observer. The
bounded, i.e., there exist an unknown positive constant 𝜏̃ ∗ magnitude of load torque estimation error |𝜏 − 𝜏|, ̂ which is
such that |𝜏|
̃ ≤ 𝜏̃ ∗ . expected to converge to zero and is also indicated as proof
in (26). The maximum bound of estimator error can be kept
Our aim is to design internal state observer so that the much lesser that of the actual disturbance. Thus, the control-
estimator error converges to origin asymptotically. This can ler requires lesser gain which results in chattering reduction.
be achieved by taking time derivative of (22) followed by
Assumption 2 and using (21)
( ) 5 Simulation Results and Discussion
𝜏̃̇ = − Q̇ + l𝜔̇ (23)
By substituting (1), (22) and (21), respectively, the load To validate the efficiency of proposed control methodologies
estimation error dynamics can be written as under uncertain load torque, the simulations were performed
( ) in MATLAB/Simulink. Two different cases were studied.
3p𝜆f The initial condition of PMSM was set to be 𝜔(0) = 0,
𝜏̃̇ = − Q̇ − l iq − b𝜔 − l𝜔 + lQ + l𝜏̃ (24)
2J iq (0) = 0, and id (0) = 0. Moreover, to observe the exact
performance of disturbance estimator, the initial condition
Let the internal state of observer to be designed as of the load torque estimator is taken to be zero. The block
( ) diagram of the overall control system is shown in Fig. 1,
3p𝜆f which includes disturbance observer that updates SMC con-
Q̇ = lQ − l iq − b𝜔 − l𝜔 (25)
2J trol law by providing instantaneous values of applied load
torque. The tuning parameters of both the control strategies
Back-substitute (25) in (24) yields are listed in Table 2. By comparing the gains requirement
we can conclude that DOSMC require less gain than con-
𝜏̃̇ = l𝜏̃ (26)
ventional SMC, which results in less chattering.

13
Journal of Electrical Engineering & Technology

Fig. 1  Overall diagram of proposed DOSMC control methodology

Fig. 3  Tracking error between actual and reference speed

Table 2  Designed feedback and Tuning Values
observer gains parameter
SMC DOSMC

k1 248.4 209.3
k2 30 25
k3 2 2
k4 25 25
c1 20 20
c2 1 1
l – − 10

Fig. 4  Load torque perturbation and estimation

nominal performance of control system due to update law

with undershoot of 3.5 rad/s due to drastic change in load
torque, that can trackback the speed trajectory in 0.5 s as
shown in zoom view of Fig. 2, which is stated as evidence
in (29). Consequently, the resulting speed tracking error for
both the proposed controllers are plotted in Fig. 3. The pro-
posed DOSMC law accurately tracks the reference speed
Fig. 2  Speed tracking under load torque variations thus converging tracking error to zero very quickly while
the conventional SMC cannot compensate the mismatch
disturbances of load torque and causing a steady error of
Case 01 The desire speed trajectory is chosen as expo- 6.48 rad/s that do not converges back to zero. The applied
nential signal. Initially, no load is coupled with shaft of load torque, observer response and observer error are shown
the motor, suddenly at time t = 2 s a constant load torque in Figs. 4 and 5 individually, for the purpose to illustrate
of TL = 0.1 Nm is applied. It is evident from Fig. 2 that adaptation performance where estimation error approaches
speed tracking of the rotor for DOSMC is same as that to zero.
of the conventional SMC in absence of disturbances. But Case 02 In this case the desire speed trajectory is formed
when the mismatch uncertainty occurs at t = 2 s, the rotor by combining two sinusoidal signals having different fre-
speed cannot track the desire trajectory causing a sever quencies. Speed trajectory tracking and load torque distur-
steady-state error of 6.48 rad/s. While DOSMC retains the bances compensation are studied in detail under uncertain

13
 Journal of Electrical Engineering & Technology

Fig. 5  Load torque estimation error Fig. 7  Speed tracking error profile

(a) conventional SMC response

Fig. 8  Load torque disturbances and estimation

sinusoidal reference speed whereas DOSMC suppress load

torque variations and regulate the speed at desired trajectory.
The corresponding speed tracking error profiles for both
the controllers are shown in Fig. 7. It is clear that DOSMC
can accurately track the desired speed. The recovery time
to minimize tracking error is 0.5 s. Contrary, conventional
SMC cannot counteract load torque disturbances. To validate
the performance of designed disturbance observer, the load
torque estimation is shown in Fig. 8. The load torque dis-
turbance estimation error is shown in Fig. 9. Initially motor
is running at no load, a sudden load of 0.1 Nm is coupled
(b) DOSMC response between 1 and 4 s, 0.3 Nm between 4 and 8 s and 0.2 Nm
between 8 and 14 s. It is very obvious that when PMSM
Fig. 6  Speed tracking comparison for sinusoidal reference trajectory is suddenly subjected to several sudden changes in applied
load torque, the designed disturbance observer exactly esti-
mates perturbations in load thus retrieve nominal speed
environment. Figure 6 plots desire and actual angular speed performance.
of PMSM under load torque perturbations to examine track- The nominal performance recovery of both the con-
ing performance. It can be seen that SMC fails to track the trollers are presented in Table 3 while examining the two

13
Journal of Electrical Engineering & Technology

References
1. Leonhard W (2001) Control of electrical drives. Springer, New
York
2. Kim IN, Son YI (2017) A modular disturbance observer-based
cascade controller for robust speed regulation of PMSM. J
Electr Eng Technol 12(4):1663–1674
3. Ang KH, Chong G, Li Y (2005) PID control system analy-
sis, design, and technology. IEEE Trans Control Syst Technol
13(4):559–576
4. Saleem Omer, Omer Urwa (2017) EKF-based self-regulation of
an adaptive nonlinear PI speed controller for a DC motor. Turk
J Electr Eng Comput Sci 25(5):4131–4141
5. Liu Heng, Li Shenggang, Cao Jinde, Li Guanjun, Alsaedi
Ahmed, Alsaadi Fuad E (2017) Adaptive fuzzy prescribed per-
formance controller design for a class of uncertain fractional-
order nonlinear systems with external disturbances. Neurocom-
puting 219:422–430
Fig. 9  Load torque disturbances estimation error 6. Sung-Kwun Oh, Roh Seok-Beom (2010) The design of fuzzy
controller based on genetic optimization and neurofuzzy net-
works. J Electr Eng Technol 5(4):653–665
7. Kim W, Chen X, Lee Y, Chung CC, Tomizuka M (2018) Dis-
Table 3  Performance analysis of SMC and DOSMC crete-time nonlinear damping backstepping control with observ-
ers for rejection of low and high frequency disturbances. Mech
Parameter Case 1 Case 2 Syst Signal Process 104(1):436–448
8. Liu C, Zhang W (2017) Adaptive backs-tepping speed control
SMC DOSMC SMC DOSMC
for IPMSM with uncertain parameters. In: Proceedings of IEEE
Max. recovery time (sec) ∞ 0.5 ∞ 0.8 International Conference on Automation, Macau, China
9. Peng Chao-Chung, Li Yunze, Chen Chieh-Li (2011) A robust
Max. undershoot 0 3.4 0 6.8
integral type backstepping controller design for control of
Steady state error (rad) 6.45 0.01 19 0.01 uncertain non-linear systems subject to disturbance. Int J Inno-
vative Comput Inf Control 7(5):2543–2560
10. Lascu C, Jafarzadeh S, Fadali MS, Blaabjerg F (2017) Direct
torque control with feedback linearization for induction motor
drives. IEEE Trans Power Electron 32(3):2072–2080
different cases in simulation results. Based on the tabu- 11. Lin C-K, Liu T-H, Yang S-H (2008) Nonlinear position con-
lated data, it can be claimed that DOSMC is more pow- troller design with input–output linearisation technique for an
erful control tool that vanishes the effect of load torque interior permanent magnet synchronous motor control system.
disturbances and retains the nominal speed performance IET Power Electron 1(1):14–26
12. Khan Owais, Pervaiz Mahmood, Ahmad Ejaz, Iqbal Jamshed
quickly. (2017) On the derivation of novel model and sophisticated
control of flexible joint manipulator. Revue Roumaine des
Sciences Techniques-Serie Electrotechnique et Energetique
62(1):103–108
13. Shansi Zhang, Shuyuan Ma, Weiming Wang (2018) Sliding
6 Conclusion mode control based on disturbance observer for magnetic levi-
tation positioning stage. J Electr Eng Technol 13(5):103–2124
In this paper, SMC and DOSMC have been presented to 14. Oveisi Atta, Nestorovićwais Tamara (2016) Robust observer-
control speed of PMSM and mitigate load torque distur- based adaptive fuzzy sliding mode controller. Mech Syst Signal
Process 76:58–71
bances. The designed disturbance observer updates the 15. Das Buddhadeva, Mhaskar Prashant (2018) Adaptive output-feed-
sliding mode surface and control input such that trajec- back Lyapunov-based model predictive control of nonlinear pro-
tory motion along the sliding manifold can drive the speed cess systems. Int J Robust Non-linear Control 28(5):1597–1609
of PMSM to the reference speed in the presence of load 16. Ahmad S, Ali N, Ayaz M, Ahmad E (2017) Design of robust
fault detection filter using algorithm for a class of LTI systems.
torque disturbances. In the simulation results, the con- In: Proceedings of IEEE International Conference on Emerging
ventional SMC performance degrades with the increase Technologies, Islamabad, Pakistan
in load torque. The proposed control scheme has dem- 17. Errouissi Rachid, Ouhrouche Mohand, Chen Wen-Hua, Trzynad-
onstrated effectiveness, trajectory tracking performance lowski Andrzej M (2012) Robust nonlinear predictive controller
for permanent-magnet synchronous motors with an optimized cost
and disturbance rejection capability. As compared with function. IEEE Trans Ind Electron 59(7):2849–2858
conventional SMC, the proposed controller technique has 18. Alam Waqar, Mehmood Adeel, Ali Khurram, Javaid Usman,
demonstrated two remarkable features that includes fast Alharbi Soltan, Iqbal Jamshed (2018) Non-linear control of a
recovery from disturbance and the designed gains are less flexible joint robotic manipulator with experimental validation.
Strojniški vestnik-J Mech Eng 64(1):47–55
than load torque estimator error rather than disturbance.

13
 Journal of Electrical Engineering & Technology

19. Mobayen Saleh, Baleanu Dumitru, Tchier Fairouz (2017) Second- Waqar Alam received his BSc
order fast terminal sliding mode control design based on LMI for a degree in electronic engineering
class of non-linear uncertain systems and its application to chaotic from UET, Peshawar in 2014.
systems. J Vib Control 23(18):2912–2925 Then he completed his master in
20. Ali N, Alam W, Rehman AU, Pervaiz M (2017) State and dis- electrical engineering from
turbance observer based control for a class of linear uncertain COMSATS university, Islama-
systems. In: Proceedings of IEEE Frontiers of Information Tech- bad in 2014. Currently, he is a
nology Conference, Islamabad, Pakistan Ph.D. scholar in COMSATS uni-
21. Jeong Min-Gil, Choi Ho-Lim (2015) Switching control of elec- versity. His research activity is
tromagnetic levitation system based on jacobian linearization and focused on nonlinear control
input–output feedback linearization. Trans Korean Inst Electr Eng schemes.
64(4):578–585
22. Liu Yingying, Jinpeng Yu, Haisheng Yu, Lin Chong, Zhao Lin
(2017) Barrier Lyapunov functions-based adaptive neural control
for permanent magnet synchronous motors with full-state con-
straints. IEEE Access 5:10382–10389
23. Wibowo WK, Jeong S-k (2013) Genetic algorithm tuned PI con- Hamid Maqsood received the
troller on PMSM simplified vector control. J Central South Univ B.S. degree from University of
20(11):3042–3048 Engineering and Technology,
Peshawar, Pakistan in 2015. He
Publisher’s Note Springer Nature remains neutral with regard to is currently pursuing his M.S.
jurisdictional claims in published maps and institutional affiliations. degree in control science and
engineering at School of Auto-
mation, Northwestern Polytech-
nical University, Xi’an, Shaanxi,
Nihad Ali received B.S. degree China. His research interests
in electronic and M.S. degree in include sliding mode control and
control engineering from Uni- its application in Unmanned Air
versity of Engineering and tech- vehicles (UAV) and robots.
n o l o g y, P e s h a w a r a n d
COMSATS university, Islama-
bad, Pakistan, in 2014 and 2017,
respectively. He is currently
working toward the Ph.D. degree
in control science and engineer-
ing at the Department of Auto-
mation, Shanghai Jiao Tong Uni-
versity, Shanghai, China. His
research interests include sliding
mode control, adaptive control
and robust control.

Ateeq Ur Rehman received the

B.Sc. degree in electronics engi-
neering from University of Engi-
neering and Technology, Pesha-
wa r, Ab b ot a b a d c a m p u s ,
Pakistan, in 2013, the M.Sc.
degree in Electrical Engineering
from Comsats University, Islam-
abad campus, Pakistan, in 2017.
Currently, he is enrolled in Ph.D.
Electrical Engineering at Paki-
stan Institute of Engineering and
Applied Sciences, Islamabad,
Pakistan. His research interests
include consensus control, non-
linear systems, and adaptive
control.

13