International Journal of Automation and Computing 8(2), May 2011, 215-220

DOI: 10.1007/s11633-011-0576-2

Adaptive Terminal Sliding Mode Control for Rigid

Robotic Manipulators
Mezghani Ben Romdhane Neila Damak Tarak
Industrial Processes Control Unit, National Engineering School of Sfax, University of Sfax, BP 1173 3038, Sfax Tunisia

Abstract: In order to apply the terminal sliding mode control to robot manipulators, prior knowledge of the exact upper bound
of parameter uncertainties, and external disturbances is necessary. However, this bound will not be easily determined because of
the complexity and unpredictability of the structure of uncertainties in the dynamics of the robot. To resolve this problem in robot
control, we propose a new robust adaptive terminal sliding mode control for tracking problems in robotic manipulators. By applying
this adaptive controller, prior knowledge is not required because the controller is able to estimate the upper bound of uncertainties
and disturbances. Also, the proposed controller can eliminate the chattering effect without losing the robustness property. The
stability of the control algorithm can be easily verified by using Lyapunov theory. The proposed controller is tested in simulation on a
two-degree-of-freedom robot to prove its effectiveness.

Keywords: Terminal sliding mode, sliding mode control, adaptive control of robot, robust control, Lyapunov method.

1 Introduction troller depends on the upper bound of parameter uncertain-

ties. Unfortunately, due to the complexity of the structure
The study of the control of rigid robotic manipulators has of uncertainties in the dynamics of robotic manipulators,
attracted growing interest in the last decade both for sci- such bound will not be easily obtained. An adaptive termi-
entific investigations and industrial needs. In fact, robotic nal sliding mode control is proposed in [13, 14] to estimate
manipulators play an important role in the industry by pro- the upper bound of uncertainties. However, the controls
viding lower production cost, enhanced precision, quality are discontinuous and in [13] five parameters must be ad-
and productivity while having greater flexibility than spe- justed. Also, an adaptive terminal sliding mode control is
cialized machines[1] . proposed in [17], where less parameters are estimated with
The control of rigid manipulators faces significant diffi- a finite time convergence. In this paper, we propose a new
culties such as highly nonlinear, highly time-varying, and robust adaptive terminal sliding mode control for tracking
highly coupled dynamic behavior. Moreover, there always problems of rigid robotic manipulators. With this control,
exists uncertainty in the system model such as external dis- a finite time convergence of the error is guaranteed and a
turbances, parameter uncertainty, and so on, which cause prior knowledge of parameter uncertainty and disturbances
unstable performance of the robotic system[2] . is not needed because the proposed controller can estimate
So far, sliding mode control is widely applied in the con- the upper bound of these uncertainties. Also, the proposed
trol of rigid robotic manipulators, thanks to its simplicity controller eliminates the chattering effect without losing the
and robustness properties[3−6] . The sliding mode control robustness property and the precision.
is characterized by robustness to parameter variations and This paper is organized as follows. The robot model is
insensitivity to disturbance[7−9] , and it has been known as presented in Section 2. A continuous terminal sliding mode
a useful strategy to deal with uncertain systems. The basic is exposed in Section 3. In Section 4, the proposed adaptive
idea of the sliding mode control is to drive and maintain terminal sliding mode control is elaborated. The stability of
the system trajectory on a sliding surface designed a pri- this control is proved by the Lyapunov theory. Simulation
ori in the state space. When the sliding mode is achieved, results are given in Section 5. Finally, conclusions are given
the system dynamics is described by the dynamics of the in Section 6.
surface, and then it becomes insensible to uncertainties sat-
isfying the matching condition. However, the sliding mode 2 Robot model
control guarantees only an asymptotic convergence. The
terminal sliding mode control has been proposed to have a Consider the dynamics of an n-link rigid robotic manipu-
fast finite time convergence[10] , and it has been applied to lator described by the following second-order nonlinear vec-
robotic manipulators[11−16] . In [11, 12] a robust multi-input tor differential equation[17]
multi-output (MIMO) terminal sliding mode controller is
M (q)q̈ + C(q, q̇) + G(q) = u + d(t) (1)
proposed for robotic manipulators. A finite time conver-
gence is guaranteed and a reduced gain of the terminal slid- where q is an n-dimensional vector of joint angles, M (q)
ing mode controller is obtained with respect to high gain is the n × n inertia matrix, C(q, q̇) is the Coriolis and
of linear sliding mode controller. To reduce the chatter- centrifugal terms, G(q) is the gravitational torque, u is
ing, the authors used a boundary layer. The proposed con- the n-dimensional vector of input torque, and d(t) is the
n-dimensional vector of the bounded input disturbance,
Manuscript received March 8, 2010; revised August 30, 2010 d(t) < d1 where d1 > 0.
216 International Journal of Automation and Computing 8(2), May 2011

Because of modeling error, parameter variations and un- We can use the equivalent control method for this goal.
known load, it is assumed that the dynamic model of The equivalent control method is used to determine the
the rigid manipulator (1) presents uncertainty. Therefore, system trajectory on the sliding surface. When the initial
M (q), C(q, q̇), and G(q) can be written as condition of the system is not on the sliding surface, the
control is the sum of a low-frequency control that is the
M (q) = M0 (q) + ΔM (q) (2) equivalent control ueq and a high frequency control Δu[22] .
The equivalent control is used to maintain the movement
C(q, q̇) = C0 (q, q̇) + ΔC(q, q̇) (3)
of the system on the sliding surface and Δu is a discontin-
G(q) = G0 (q) + ΔG(q). (4) uous control that drives the system trajectory to reach the
sliding surface. The equivalent control can be determined,
Then, from (2)–(4), (1) can be written in the following
in the absence of disturbances and uncertainties, from
form
Ṡ = 0.
M0 (q)q̈ + C0 (q, q̇) + G0 (q) = u + ρ(t)
where ρ(t) is defined as The expression of the equivalent control is then

ρ(t) = −ΔM (q) − ΔC(q, q̇) − ΔG(q) + d(t). a a −1

ueq = M0 (q)(q̈d − Cdiag{e1b }) + C0 (q, q̇) + G0 (q). (8)
b
Some assumptions are used as below:
Assumption 1. The norm of inertia matrix M (q) is up- The discontinuous term Δu is[23]
per bounded by a positive number α0 [17]
 T T
M (q) < α0 . S M0 (q)−1
Δu = − T ×
Assumption 2. The following inequality is verified S M0 (q)−1 2
  
SM0 (q)−1  b0 + b1 q + b2 q̇2 .
C(q, q̇) + G(q) < β0 + β1 q + β2 q̇2
However, the discontinuous control Δu can result in the
where β0 , β1 , and β2 are positive numbers. It is shown in chattering problem, and to overcome this undesirable effect
[17] that the uncertainty is input related and if the control Δu can be replaced by the following expression[23]
input does not contain the acceleration signal, the system
uncertainty will be bounded by a positive function of the
position and velocity measurements in the following form Δu1 =
⎧
⎪
⎪ (S T M0 (q)−1 )T
ρ(t) < b0 + b1 q + b2 q̇2 . (5) ⎪
⎪ − [SM0 (q)−1 ×
⎪
⎪ S T M0 (q)−1 2
⎪
⎪
This bounded property has been used by some re- ⎨ (b0 + b1 q + b2 q̇2 )], if S T M0 (q)−1   δ
searchers of [15–18]. ⎪
⎪
⎪ (S T M0 (q)−1 )T
⎪
⎪ − [SM0 (q)−1 ×
⎪
⎪ δ2
3 Continuous terminal sliding mode ⎪
⎩
(b0 + b1 q + b2 q̇2 )], if S T M0 (q)−1  < δ
control
where δ > 0.
The trajectory tracking control of the robot manipulator The control applied to the rigid robotic manipulator is
can be formulated as follows: let qd ∈ Rn be a given twice
differentiable desired trajectory, and define the tracking er- u = ueq + Δu1 .
ror as e1 = q − qd ; the control objective is to find a feedback
control law u such that the manipulator output q tracks the The continuous terminal sliding mode control eliminates
desired trajectory qd , the tracking error converges to zero in the chattering phenomenon at the cost of robustness prop-
finite time. For this purpose, consider the terminal sliding erty, and this controller depends on the upper bound of the
surface[16−21] a
uncertainties and disturbances which is difficult to deter-
S = e2 + Ce1b (6) mine.
where e2 = q̇ − q˙d , C = diag{c1 , · · · , cn }, and a and b are
odd integers satisfying 0 < a < b. 4 Robust adaptive terminal sliding
The dynamic error corresponding to (1) is mode control
⎧
⎪ ė = e2 Applying the terminal sliding mode control for rigid
⎨ 1
ė2 = −q̈d − M0 (q)−1 (C0 (q, q̇) + G0 (q))+ (7) robotic manipulators needs the knowledge of the upper
⎪
⎩ bound of uncertainties and disturbances in advance. How-
−1 −1
M0 (q) u + M0 (q) ρ(t). ever, in the case of manipulators, the complexity and un-
After choosing the sliding surface, the second step is to predictability of the structure of uncertainties may partic-
determine a control law satisfying the sliding condition ularly cause certain difficulties in determining this bound.
Besides, to eliminate the chattering effect, an increase of
S T Ṡ < 0. the parameter δ will generate the loss in robustness. To
 M. B. R. Neila and D. Tarak / Adaptive Terminal Sliding Mode Control for Rigid Robotic Manipulators 217

overcome these problems, we propose a new robust adap- ˙

Simplifying and substituting b̂i by the expression defined
tive terminal sliding mode controller for the robotic ma- by (10), we obtain
nipulators described by (1), where the disturbance and the
uncertainty satisfy (2)–(4), in order to estimate the bound V̇ =S T M0 (q)−1 ρ(t)−
of uncertainties and external disturbances online. Using SM0 (q)−1 (b̂0 + b̂1 q + b̂2 q̇2 )−
the sliding surface (6), the adaptive terminal sliding mode
control is proposed as follows SM0 (q)−1 (b̃0 + b̃1 q + b̃2 q̇2 ) =
S T M0 (q)−1 ρ(t)−
u = ueq + Δu2 (9) SM0 (q)−1 (b0 + b1 q + b2 q̇2 ) 
where ueq is defined in (8) and Δu2 has the following ex- SM0 (q)−1 (ρ(t) − (b0 + b1 q + b2 q̇2 ) < 0
pression . 

Δu2 =
⎧ 5 Simulation results
⎪
⎪ (S T M0 (q)−1 )T
⎪
⎪ − [SM0 (q)−1 × The performance of the proposed controller is tested via
⎪
⎪ S T M (q)−1 2
0
⎪
⎪
⎨ (b̂0 + b̂1 q + b̂2 q̇2 )], if S T M0 (q)−1   δ
simulation on a two-degrees-of-freedom robot described by
the following model[17]
⎪
⎪ (S T M0 (q)−1 )T
⎪
⎪ − [SM0 (q)−1 ×
⎪
⎪ δ2
⎪
⎪     
⎩
(b̂0 + b̂1 q + b̂2 q̇2 )], if S T M0 (q)−1  < δ M11 (q) M12 (q) q̈1
+
C1 (q, q̇)
+
M12 (q) M22 (q) q̈2 C2 (q, q̇)
where b̂0 , b̂1 , and b̂2 are the adaptive variables for b0 , b1 ,      
G1 (q) u1 d1 (t)
and b2 defined in (5). The adaptation laws are = +
G2 (q) u2 d2 (t)
˙
b̂0 = x0 SM0 (q)−1  where
˙
b̂1 = x1 SM0 (q)−1 q (10) M11 (q) = (m1 + m2 )L21 + m2 L22 + 2m2 L1 L2 cos(q2 ) + J1

M12 (q) = m2 L22 + m2 L1 L2 cos(q2 )

˙
b̂2 = x2 SM0 (q)−1 q̇2
M22 (q) = m2 L22 + J2
where x0 , x1 , and x2 are arbitrary positive constants.
C1 (q, q̇) = −m2 L1 L2 sin(q2 )q̇12 − 2m2 L1 L2 sin(q2 )q̇1 q̇2
Theorem 1. If the control law (9), with the sliding sur-
face (6) and the adaptation law (10), is applied to the non- C2 (q, q̇) = m2 L1 L2 sin(q2 )q̇2
linear uncertain system defined by (7), the error converges
to zero in finite time. G1 (q) = (m1 + m2 )L1 cos(q2 ) + m2 L2 cos(q1 + q2 )
Proof. Let us consider the following positive definite
G2 (q) = m2 L2 cos(q1 + q2 ).
function as a Lyapunov function candidate
The nominal values of m1 and m2 are assumed to be[17]
1 T 1

2
V = S S+ x−1 2
i b̃i m10 = 0.4 kg, m20 = 1.2 kg
2 2 i=0
and we suppose that we have an uncertainty on masses of
where b̃i = bi − b̂i , i ∈ {0, 1, 2}. the order ±10% (see Figs. 1 and 2). The other system pa-
Differentiating V with respect to time and using the con- rameters are assumed to be known[17] :
trol law (9) for S T M0 (q)−1   δ yields
L1 = 1 m, L2 = 0.8 m
T −1
V̇ =S [−q̈d − M0 (q) (C0 (q, q̇) + G0 (q))+ J1 = 5 kg·m, J2 = 5 kg.
M0 (q) −1
(ueq + Δu2 ) + M0 (q) −1
ρ(t)+ The disturbance vector is d(t) = [d1 (t) d2 (t)]T , where
a a −1
2
˙
Cdiag{e1b }e2 ] − x−1
i b̃i b̂i
d1 (t) = 0.2 sin(3t) + 0.02 sin(26πt)
b i=0
d2 (t) = 0.1 sin(2t) + 0.01 sin(26πt).
In this example, the initial values of the system are se-
(S T M0 (q)−1 )T
V̇ =S T M0 (q)−1 ρ(t) − [S T M0 (q)−1 × lected as
S T M0 (q)−1 2

2
˙ [q1 (0) q2 (0)]T = [0.8 0.9]T
SM0 (q)−1 (b̂0 + b̂1 q + b̂2 q̇2 )] − x−1
i b̃i b̂i
i=0 [q̇1 (0) q̇2 (0)]T = [0 0]T .
218 International Journal of Automation and Computing 8(2), May 2011

We desire the two articulations track, respectively, the

following desired angular positions[17]

7 7
qd1 = 1.25 − exp(−t) + exp(−4t)
5 20

7 7
qd2 = 1.4 − exp(−t) + exp(−4t).
5 20
The chosen surface parameters are a = 5, b = 7, and
c1 = c2 = 2, and the initial conditions of the upper bound
of the uncertainty are b00 = 5, b10 = 14, and b20 = 2. The
results obtained for the continuous terminal sliding mode
control and the adaptive terminal sliding mode control are
given by Figs. 3 and 4. In order to have a small bound-
ary layer around the sliding surface to have a more robust
system, we have chosen small δ equal to 0.0005. However,
Fig. 1 Variation of mass m1
the control law is not totally continuous, and it presents
a high frequency commutation at the instant 1 s which is
undesirable in practice. Besides, the controller depends on
the upper bound of uncertainties and disturbances. By the
application of the new adaptive terminal sliding mode con-
trol, these problems are solved conserving the same param-
eter δ. In fact, the proposed control is continuous without
losing the robustness property, and the parameters of the
upper bound are estimated online. Figs. 5–7 present these
parameters. These results show a finite time convergence of
the upper bound parameters. In the presence of large un-
certainty, the terminal sliding mode control can give large
tracking error which is not the case in adaptive terminal
sliding mode control. In fact, according to the adaptive laws
(10), the control is adjusted to have a very small tracking
error. Therefore, the effect of the uncertainty can be elimi-
nated. The obtained results present some improvements in
convergence time of the error and control amplitude com-
Fig. 2 Variation of mass m2 pared to the results of [17].

Fig. 3 Tracking of the first joint with terminal sliding mode control (dotted line) and adaptive terminal sliding mode control (solid
line)

Fig. 4 Tracking of the second joint with terminal sliding mode control (dotted line) and adaptive terminal sliding mode control (solid
line)
 M. B. R. Neila and D. Tarak / Adaptive Terminal Sliding Mode Control for Rigid Robotic Manipulators 219

approach. This adaptive algorithm is used to estimate the

bounds of uncertainties and external disturbances. With
this controller, a finite time convergence of the error is guar-
anteed and the knowledge of the upper bound of the distur-
bances and uncertainties is not necessary. The simulation
results show that the algorithm can estimate this bound on-
line and can assure a good performance. In fact, the error
converges to zero in finite time and the proposed control
is robust to uncertainties and disturbances. Also, the pro-
posed control has eliminated the chattering phenomenon
without losing the robustness property and precision.

Acknowledgement
The authors would like to thank the members of the In-
dustrial Processes Control Unit.
Fig. 5 Estimated parameter b̂0
References
[1] R. Kelly, V. Santibanez, A. Loria. Control of Robot Manip-
ulators in Joint Space, Berlin, Germany: Springer-Verlag,
2005.

[2] Y. Z. Guo, P. Y. Woo. Adaptive fuzzy sliding mode control

for robotic manipulators. In Proceedings of the 42nd IEEE
Conference on Decision and Control, IEEE, Maui, USA,
vol. 3, pp. 2174–2179, 2003.

[3] J. M. Yang, J. H. Kim. Sliding mode control for trajec-

tory tracking of nonholonomic wheeled mobile robots. IEEE
Transactions on Robotics and Automation, vol. 15, no. 3,
pp. 578–587, 1999.

[4] Q. H. Nguyen, E. Kreuzer. A robust adaptive sliding

mode controller for remotely operated vehicles. Technische
Mechanik, vol. 28, no. 3–4, pp. 185–193, 2007.

[5] U. Vadim, G. Jurgen, J. X. Shi. Sliding Mode Control in

Electro-mechanical Systems, UK: CRS Press, 1999.
Fig. 6 Estimated parameter b̂1
[6] F. Li, H. L. Xie. Sliding mode variable structure control for
visual servoing systems. International Journal of Automa-
tion and Computing, vol. 7, no. 3, pp. 317–323, 2010.

[7] A. Benchaib, M. Tadjine, A. Rachid. Sliding mode control

of an induction motor with unknown load: Application on
a digital-signal-processor-based system. International Jour-
nal of Systems Science, vol. 30, no. 8, pp. 849–863, 1999.

[8] M. J. Elbrous. Robust sliding mode controllers design tech-

niques for stabilization of multivariable time-delay systems
with parameter perturbations and external disturbances.
International Journal of Systems Science, vol. 36, no. 7,
pp. 433–444, 2005.

[9] B. Bandyopadhyay, A. G/E. Alber, S. Janardhanan, V.

Sreeram. Sliding mode control design via reduced order
model approach. International Journal of Automation and
Computing, vol. 4, no. 4, pp. 329–334, 2007.

[10] X. H. Yu, J. X. Xu. Variable Structure Systems: Towards

Fig. 7 Estimated parameter b̂2 the 21st Century, Berlin, Germany: Springer-Verlag, 2002.

[11] Z. H. Man, A. P. Paplinski, H. R. Wu. A robust MIMO ter-

6 Conclusions minal sliding mode control scheme for rigid robotic manip-
ulators. IEEE Transactions on Automatic Control, vol. 39,
In this paper, we have described the design of a new ro- no. 12, pp. 2464–2469, 1994.
bust adaptive terminal sliding mode controller for the track-
[12] K. B. Park, J. J. Lee. Comments on a robust MIMO ter-
ing problem of the rigid robotic manipulators. The main minal sliding mode control scheme for rigid robotic manip-
feature of this design is that it combines the terminal slid- ulators. IEEE Transactions on Automatic Control, vol. 41,
ing mode control with a boundary layer and the adaptive no. 5, pp. 761–762, 1996.
220 International Journal of Automation and Computing 8(2), May 2011

[13] Z. H. Man, O. Mike, X. H. Yu. A robust adaptive terminal [22] B. Ahcene, B. Fares, L. Salim. A neuro-fuzzy-sliding mode
sliding mode control for rigid robotic manipulators. Journal controller using nonlinear sliding surface applied to the cou-
of Intelligent and Robotic Systems, vol. 24, no. 1, pp. 23–41, pled tanks system. International Journal of Automation and
1999. Computing, vol. 6, no. 1, pp. 72–80, 2009.
[14] P. G. Keleher, R. J. Stonier. Adaptive terminal sliding [23] M. B. R. Neila, D. Tarak. Terminal sliding mode control for
mode control of a rigid robotic manipulator with uncertain robotic manipulators with uncertain dynamics and bounded
dynamics incorporating constraint inequalities. ANZIAM
disturbance. In Proceedings of the 10th International Con-
Journal, vol. 43, pp. 102–153, 2001. ference on Sciences and Techniques, pp. 1641–1652, 2009.
[15] S. H. Yu, X. H. Yu, R. Stonier. Continuous finite time con-
trol for robotic manipulators with terminal sliding modes.
In Proceedings of the 6th International Conference of In- Mezghani Ben Romdhane Neila
formation Fusion, IEEE, vol. 2, pp. 1433–1440, 2003. graduated from National Engineering
[16] Y. C. Huang, T. Z. S. Li. Fuzzy terminal sliding-mode con- School of Sfax (ENIS), Tunisia in 2002.
troller for robotic manipulators. In Proceedings of IEEE She received the master degree in auto-
International Conference on Mechatronics, IEEE, Taipei, matic and industrial data processing from
Taiwan, pp. 858–863, 2005. ENIS in 2005. She has joined the ENIS as
an associate assistant of electric engineering
[17] Z. H. Man, X. Yu. Adaptive terminal sliding mode track- since 2005.
ing control for rigid robotic manipulators with uncertain Her research interests include sliding
dynamics. JSME International Journal, Series C, vol. 40, mode control, adaptive nonlinear control,
no. 3, pp. 493–502, 1997. and control of robotic manipulator.
E-mail: [email protected] (Corresponding author)
[18] C. Abdallah, D. Dawson, P. Dortao, M. Jamshidi. Survey
of robust control for rigid robots. IEEE Control Systems
Magazine, vol. 11, no. 2, pp. 24–30, 1991. Damak Tarak graduated from Na-
tional Engineering School of Sfax (ENIS),
[19] D. M. Wang. The design of terminal sliding controller of Tunisia in 1989. He received the  Diplome
two-link flexible manipulators. In Proceedings of IEEE In- d Etudes Approfondues in automatic con-
ternational Conference on Control and Automation, IEEE, trol from the Institut National des Sciences
Guangzhou, PRC, pp. 733–737, 2007. Appliquées de Toulouse, France in 1990
[20] X. H. Yu, Z. H. Man. On finite time mechanism: Terminal and the  Doctorat d Université degree from
sliding modes. In Proceedings of IEEE International Work- the Université Paul Sabatier de Toulouse,
shop on Variable Structure Systems, IEEE, Tokyo, Japan, France in 1994. In 2006, he obtained the
pp. 164–167, 1996. University Habilitation from the ENIS.
His research interests include disturbed parameter systems,
[21] H. Liu, J. F. Li. Terminal sliding mode control for space- sliding mode control and observers, and adaptive nonlinear con-
craft formation flying. IEEE Transactions on Aerospace and trol.
Electronic Systems, vol. 45, no. 3, pp. 835–846, 2009. E-mail: [email protected]