An Event-triggered Based Robust Control of

Robot Manipulator

Niladri Sekhar Tripathy 1

I.N. Kar 1

Kolin Paul 2

Dept. of Electrical Engineering

2 Dept. of Computer Science & Engineering,
Indian Institute of Technology Delhi,
New Delhi, INDIA
1

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

What is event-triggered control ?

Execute the control task when strictly necessary (but not

periodically) to achieve the desired goal. In event-triggered control
the input is held constant while performance is satisfactory.

Figure 1: Block diagram of event-triggered control.

And how do we know when it is necessary to execute the

Event-triggered control requires the constant

monitoring of the state to determine current performance.
What is prior assumption? Processor is available when it is
needed.
control task?

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Modelling of event-triggered state feedback control system

Linear system:
x(t)

= Ax(t) + Bu(t)

Continuous time state-feedback control law: u(t) = Kx(t)

Event-triggered control law:
u(tk ) = Kx(tk )

The time sequence {tk }kI represents aperiodic state

transmission, control input computation and actuation instant.
Sensing, computation and actuation all these three tasks are
occurring instantly at the time of event generation.
Event-triggered system is modelled as
x

= Ax + BK (x + e)

where e R n is referred as measurement error.

e(t) = x(tk ) x(t), t [tk , tk+1 ), k I
Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Conceptual block diagram of proposed event-trigger based

robust control.

Figure 2: Block diagram of proposed event-trigger based robust control

of robot manipulator.
Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Problem description and statement

Design a robust control law for uncertain manipulator

dynamics.
u(t) = Kx(t),

Control law is designed by minimizing a modied performance

functional.
The event-based state feedback control input will be
u(tk ) = Kx(tk )

The actuation instant of the control input {tk }, is determined

through a state-dependent event condition.
The event condition is derived to ensure the closed loop
stability.
Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

System uncertainty

System with uncertainty

Linear uncertain system:

x = A(p)x + Bu

where p P is an uncertain parameter vector.

Matched uncertainty:
A(p) A(p0 ) = B(p)

for any p P .
System uncertainty is assumed to be in the range space of
input matrix B .

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

System with matched and input uncertainty:

Nonlinear system having both system and input uncertainty:

x(t)

= A(x) + B(x)(u + h(x)u) + B(x)f (x)

f (x) Rm1 and h(x) Rmm are system and input uncertainty,

respectively.

Assumption 1.

A(0) = 0 and f (0) = 0 so that x = 0 is an equilibrium point.

Assumption 2.

kf (x)k fmax (x)

Assumption 3.

h(x) 0

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Nonlinear manipulator dynamics

The dynamics of robot manipulator:

M(q)
q + N(q, q)
=

where N(q, q)
= V (q, q)
+ F (q)
+ G (q). Variables q and
denote the joint variables and generalized forces. The inertia
matrix, Coriolis vector, gravity vector and friction vector are
represented by M(q), V (q, q),
G (q) and F (q)
respectively.
The state space representation of robot manipulator with
uncertainty:
   
x
q
x(t)

= 1 = 1 = Ax(t) + B(u + h(x)u) + Bf (x)

x2
q 2

0 I
I
where A =
and B =
.
0 0
0


Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

 

An Event-triggered Based Robust Control of Robot Manipu

The system and input uncertainty f (x) and h(x) are

h(x) = M(x1 )1 M0 (x1 ) I 0
f (x) = M(x1 )1 (N0 (x1 , x2 ) N(x1 , x2 ))
M0 (x1 ) and vector N0 (x1 , x2 ) represent the nominal value of
M(x1 ) and N(x1 , x2 ).
Assumption 4:

System uncertainty satises

f (x)T f (x) x T Fx

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Robust control problem

u(t) = Kx(t)
p P.

Aim: Find a state feedback control law

uncertain system is stable for any

such that the

Optimal control input for nominal

system, which minimizes the modied cost functional, is also the
robust solution of uncertain system.
Nominal dynamics and cost functional for matched uncertain
system:
Optimal control Method:

x(t)

= Ax(t) + Bu(t)
Z

(x T Fx + x T x + u T u)dt

The optimal control input is designed.

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Controller design

Step 1: The Lyapunov function V (x) should satisfy Hamilton

Jacobi Bellman (HJB) equation

minu1 R m (x T Fx + x T x + u T u + VxT (Ax + Bu)) = 0

where Vx =
Then

V
x

and u = Kx .

x T Fx + x T x + u T u + VxT (Ax + Bu) = 0

Step 2.

The optimal inputs u(t) minimizes the Hamiltonian

H(x(t), u(t), Vx ) = x T Fx + x T x + u T u + VxT (Ax + Bu)

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

which leads to
H(x(t), u(t), Vx , t)
= x T K T + VxT B = 0
u(t)

If V (x) = x T Sx , the HJB equation reduces to algebraic

Riccati equation

Step 3.

SA + AT S + F + I SBB T S = 0

The optimal control input:

u = B T Sx
Step 4. Control gain K and aperiodic state information of original
system x(tk ) are used to compute the event-based control control
law

u(tk ) = Kx(tk )
Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Triggering condition design

Positive inter-execution time (tk+1 tk = > 0).

The closed loop system must be input to state stable (ISS).
To prove this there must have an ISS Lyapunov function.
Input to state stability:

Denition
A continuous function

V : Rn R

is an ISS Lyapunov function for

the event-triggered system, if there exist class

1 , 2 , 3

and

for all

x, e

Rn

functions

and it satisfy

1 (kx(t)k) V (x(t)) 2 (kx(t)k)

V (x)x 3 (kx(t)k) + (ke(t)k)

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

The ISS condition helps to construct the event-triggering rule

in-terms of measurement error norm ke(t)k and the state
norm kx(t)k.
Triggering rule:
(ke(t)k) 6 (kx(t)k)

where (0, 1), and need to be designed.

ISS Lyapunov function:
V (x) = x T Sx

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Robust control law for event-triggered system

Uncertain manipulator dynamics with event-triggered control law:

x(t)

= Ax(t) + {B(Kx(t) + Ke(t)) + h(x)(Kx(t) + Ke(t))} + Bf (x)

Theorem
Suppose the controller gain matrices

is designed for the nominal

system by minimizing the cost functional then the uncertain

system, with event-based control law is ISS if there exist a event
occurring sequence

{tk }kI

given by

t0 = 0, tk+1 = inf {t R|t > tk kxk kek 0}

where the parameters are given by

=

min (Q)

2kSBK k

and Q = ((K T hK + K T K )T (K T hK + K T K ) + I )
Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Minimum time interval in between two consecutive events

In event-triggered control inter execution time depends on the

evolution of kek/kxk with respect to time. At tk , kek/kxk is zero
as measurement error e = 0. The next event will occur at tk+1 ,
when the kek/kxk turns to .
Theorem
The uncertain system with triggering law has strictly positive lower
bound of inter-event time

>0

and it is expressed as

q
L23 4L2 L1
2
q
=q
ln{k q
kk
k}
2
2
2
L3 4L2 L1
L3 4L2 L1 L3 L3 + 4L2 L1

2L2 + L3

L3 +

L1 = (kA + B(I + h(x))K k + kBkkF k),

L2 = kB(I + h(x))K k and L3 = L1 + L2

where

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Simulation Results

Linear system:

T
x(t)
= [x1 (t) x2 (t)
[q1 (t) q2 (t) q1 (t) q2 (t)]T ,
x3 (t) x4 (t)]
=

0 0 1 0

0 0

0 0 0 1

and B = 0 0 .
A=
0 0 0 0
1 0

0 0 0 0
0 1
System and input uncertainty:

f (x) =

562.0 + 171.6 cos x2 + 100mL + 96mL cos x2

51.2 + 85.8 cos x2 + 36mL + 48mL cos x2

51.2 + 85.8 cos x2 + 36mL + 48mL cos x2

51.2 + 36mL

48mL (2x3 x4 )x4 sin x2 + 78.4mL sin x1 + 58.8mL sin(x1 + x2 )

48mL x3 2 sin x2 + 58.8mL sin(x1 + x2 )

562.0 + 171.6 cos x2 + 100mL + 96mL cos x2

51.2 + 85.8 cos x2 + 36mL + 48mL cos x2


h(x) =

1

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

51.2 + 85.8 cos x2 + 36mL + 48mL cos x2

51.2 + 36mL

2562.0 + 2091.6 cos x2

771.2 + 1045.8 cos x2

771.2 + 1045.8 cos x2

771.2

!

1


I

An Event-triggered Based Robust Control of Robot Manipu

The mass of the load (mL ): Varies from 10 to 20 oz

Upper bound of uncertainty:

27.6
14.3
198.6 81.7
14.3
8.6
105.3 35.1

F =
198.6 105.3 1432.9 573.2.
81.7 35.1 573.2 286.6
Algebraic Riccati equation: AT P + PA + F + I PBB T P = 0.
The
 event-triggered
  control
 input u(tk ) =


4.98 1.93 x1 (tk )
36.12 11.73 x3 (tk )

.
1.93 2.42 x2 (tk )
11.73 12.40
x4 (tk )
Event-triggering instant:
t0 = 0, tk+1 = inf {t R|t > tk ke(t)k2 0.04kx(t)k2 }

Initial states: [/4, /2, 0, 0]T , simulation time=25 sec.

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

(a) Angle positions of two-link

(b) Angle velocities of two-link

manipulator for

manipulator for

mL

= 10 oz

mL

= 10 oz

Figure 3: Angle positions.

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

(a) Angle positions of two-link

(b) Angle velocities of two-link

manipulator for

manipulator for

mL

= 20 oz

mL

= 20 oz

Figure 4: Angle velocities.

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Figure 5: Control input at respective event-generating instant for

mL

20 oz.

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Conclusion

Event-triggered based robust control law for uncertain robot

manipulator.
The closed loop event-triggered system is ISS for bounded
variation of parameters.
The total number of control input computation and information
transmission are less in an event-based mechanism over
conventional continuous approach.
Feature work

The detailed analysis of design parameters are not addressed.

Self-triggered approach may be considered to solve the robust

control problem as a future work.

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Thank You

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Bibliography
K. Astrom and B. Bernhardsson, Comparison of Riemann and Lebesgue sampling for rst order
stochastic systems, 41st IEEE Conference on Decision and Control, pp. 2011-2016, 2002.
W.P.M.H. Heemels, K.H. Johansson and P. Tabuada, An introduction to event-triggered and
self-triggered control, 51st IEEE Conference on Decision and Control, pp. 3270-3285, 2012.
P. Tabuada, Event-triggered real-time scheduling of stabilizing control tasks,

on Automatic Control, vol. 52(9), pp. 1680-1685, 2007.

IEEE Transactions

Nicolas Marchand, Sylvain Durand, and Jose Fermi Guerrero Castellanos, A General formula for
event-based stabilization of nonlinear systems, IEEE Transactions on Automatic Control, vol.
58(5), pp. 1332-1337, 2013.
A. Anta, and P. Tabuada, To sample or not to sample:self-triggered control for nonlinear systems.
IEEE Transactions on Automatic Control, vol. 55(9), pp. 2030-2042, 2010.
A. Girard, Dynamic event generators for event-triggered control systems,
http://arxiv.org/pdf/1301.2182v1.pdf, 2013.
A. Eqtami, D. V. Dimarogonas and Kostas J. Kyriakopoulos, Event-triggered control for
discrete-time systems, American Control Conference, pp. 4719-4724, 2010.
M. C. F. Donkers and W. P. M. H. Heemels, Output-based event-triggered control with
guaranteed L -gain and improved and decentralized event-triggering, IEEE Transactions
Automatic Control, vol. 57(6), pp. 1362-1376, 2012.

on

Avimanyu Sahoo, Hao Xu and S. Jagannathan, Neural network-based adaptive event-triggered

control of nonlinear continuous-time systems, IEEE Multi-Conference on Systems and Control,
pp.35-40, 2013.
Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

E. D. Sontag, Input to state stability: basic concepts and results,

Theory, pp. 163-220, 2008.

Nonlinear and Optimal Control

D. Nesic and A.R. Teel, Input-to-state stability of networked control systems,

40(12), pp. 2121-2128, 2004.

Automatica, vol.

P. Tallapragada and N. Chopra, Event-triggered dynamic output feedback control for LTI
systems, 51st IEEE Conference on Decision and Control, pp. 6597-6602, 2012.
S. Trimpe and R. DAndrea, Event-based state estimation with variance-based triggering,
IEEE Conference on Decision and Control, pp. 6583-6590, 2012.
P. Tallapragada and N. Chopra, On event triggered tracking for nonlinear systems,

Transactions on Automatic Control, vol. 58(9), pp. 2343-2348, 2013.

51st

IEEE

Avimanyu Sahoo, Hao Xu and S. Jagannathan. Adaptive event-triggered control of a uncertain

linear discrete time system using measured input and output data, American Control Conference,
pp.5672 -5677, 2013

CRC press, First Indian Reprint, 2009.

Gene H, Van Loan Golub, Matrix computations, Hindustan Book Agency (India), 3rd edition,
D. S. Naidu, Optimal control systems,

2007.

E. Garcia and P. J. Antsaklis, Model-based event-triggered control for systems with quantization
and time-varying network delays, IEEE Transactions on Automatic Control, vol. 58(2), pp.
422-434, 2013.
F Lin, Andrzej W. Olbrot. An LQR approach to robust control of linear systems with uncertain
parameters, 35th IEEE Conference on Decision and Control, pp. 4158-4163, 1996.
F Lin. An optimal control approach to robust control design,
73(3), pp. 177-186, 2000.
Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

International journal of control vol.

An Event-triggered Based Robust Control of Robot Manipu

D.M. Adhyaru, I.N. Kar and M. Gopal, Fixed nal time optimal control approach for bounded
robust controller design using Hamilton Jacobi Bellman solution, IET Control Theory and
Applications. vol. 3(9), pp. 1183-1195, 2009.
D.M. Adhyaru, I. N. Kar and M. Gopal, Bounded robust control of systems using neural network
based HJB solution, Neural Comput and Applic, vol. 20(1), pp. 91-103, 2011.
F. Lin and R. D. Brandt, An optimal control approach to robust control of robot manipulators,
IEEE Transactions on Robotics and Automation, vol. 14(1), pp. 69-77, 1998.
Niladri Sekhar Tripathy, I.N. Kar and Kolin Paul, An event-triggered based robust control of robot
manipulator. 13th International Conference on Control, Automation, Robotics and Vision,
Singapore, 2014 (Accepted).
M. Xia, V. Gupta and P. J. Antsaklis.Networked state estimation over a shared communication
medium, American Control Conference, pp. 4128-4133 , 2013
C. Santos, M. Mazo, F. Espinosa, Adaptive self-triggered control of a remotely operated robot,
Advances in Autonomous Robotics, vol. 7429, pp. 61-72, 2012.
Karl Henrik Johanssona, Magnus Egerstedtb, John Lygerosa, Shankar Sastry, On the
regularization of Zeno hybrid automata, Systems & Control Letters, vol. 38, pp. 141-150, 1999.

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

Proof.
If V (x) is the Lyapunov function for (1), and (10) is the modied
cost function. Then
(1)

V (x) = VxT x(t)

VxT (Ax

+ B{(Kx + Ke) + h(x)(Kx + Ke)}

+Bf (x))


Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu

V (x) = (x T Fx + x T x + 2x T K T hKx
+2x T (K T K + K T hK )e
+(u0 + f (x))T (u0 + f (x))) + f (x)T f (x)

Using upper bound of system uncertainty (9), condition (1) &

(1) hold if
(kx(t)k) = min (Q )kx(t)k2
0

(ke(t)k) = min (Z )ke(t)k2

0

(2)
(3)

Q and Z are dened as

Q
Z

0
0

= K T h(x)K + (1 1/)I
T

= K K + K h(x)K

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

(4)
(5)

An Event-triggered Based Robust Control of Robot Manipu

The control input update is required only when the following

triggering condition is violated:
ke(t)k2 (t)kx(t)k2

where

(6)

(t) =

min (Q )
min (Z 0 )

(7)

Using (6)and (7), the time instant, at which the event has
occurred:
t0 = 0, tk+1 = inf {t R|t > tk (t)kxk2 kek2 0}

(8)

Then V (x) becomes

0
V (x) ( 1)min (Q )kx(t)k2

The uncertain event-triggered system (1) is ISS (0, 1) .

Niladri Sekhar Tripathy, I.N. Kar, Kolin Paul

An Event-triggered Based Robust Control of Robot Manipu