IJSTE - International Journal of Science Technology & Engineering | Volume 1 | Issue 11 | May 2015

ISSN (online): 2349-784X

Design of Kharitnov Polynomial based State

Feedback Controller for Rotary Inverted
Pendulum
Jim George
Assistant Professor
Department of Electrical and Electronics Engineering
Muthoot Institute of Technology and Science, Cochin, Kerala, India

Abstract
This paper deals with design of State feedback controller for Rotary Inverted Pendulum based on Kharitnov polynomial
approach . Rotary Inverted Pendulum (RIP) is a nonlinear and unstable system. The key idea here is to stabilize the position of
the pendulum at inverted position for the rotary base. This paper deals with mathematical modeling , design of state feedback
controller and stability analysis for the system specified.
Keywords: Rotary Inverted Pendulum (RIP), State Feedback Controller, Kharitnov Polynomial, Stability Analysis
________________________________________________________________________________________________________

I. INTRODUCTION
A Rotary Inverted Pendulum consists of a pendulum attached to a rotary arm driven by a motor. The aim of this paper is to
design a State Feedback Controller for this system which is here considered as Single Input Single Output System (SISO
system). In this paper design of State feedback swing up controller is dealt with whose stabilization point is the inverted position
of the pendulum while the base is moving in a circular path.

II. ROTARY INVERTED PENDULUM

Fig. 1: Rotary Inverted Pendulum

Fig. 2: Side View of the Pendulum in Motion

A RIP consist of a) rotating arm b) Pendulum c) Motor for deriving the rotating arm

All rights reserved by www.ijste.org

217

Design of Kharitnov Polynomial based State Feedback Controller for Rotary Inverted Pendulum
(IJSTE/ Volume 1 / Issue 11 / 035)

III. MATHEMATICAL MODELLING

A. Modelling of Rotating arm and Pendulum:
Let the pendulum be considered as lump mass at half the length of the pendulum.
The side view of the pendulum on motion can be shown as in Fig 2.
L is the length to pendulums centre of mass (half the pendulum length), m is the mass of the pendulum,
r is the half the length rotating arm, is the pendulum deflection, is the servo load gear angle
From fig 3, there are two components for velocity of the pendulum lumped mass
( )
( )
Velocity of the rotating arm:

Taking the x and y components:

( )
( )

(1)
(2)
(3)
(4)

B. System Dynamic Equations:

Potential Energy is only due to gravity.
(5)
Kinetic Energy arises from the moving hub, the velocity of point mass in the x-direction, the velocity of the point mass in
the y-direction and the rotating pendulum about its center of mass:
(6)
The moment of inertia of a rod about its centre of mass is:
(7)
Taking:
,
(8)

(9)
(10)
(11)

The Lagrangian is given by:

(12)
(13)

( )

(14)

( )

(15)

Output torque on the load of the motor is:

(

(16)

C. State Space Model of RIP System:

The state space model of one sided rotary inverted pendulum can be deduced from the above Expressions as shown below,

[ ]

[ ]
]

Where:

(17)
[
,

]
,

All rights reserved by www.ijste.org

218

Design of Kharitnov Polynomial based State Feedback Controller for Rotary Inverted Pendulum
(IJSTE/ Volume 1 / Issue 11 / 035)

IV. STATE MODEL OF RIP SYSTEM

Specificatio
ms

Mass
(m)

Length (L)

Pendulum

100g

20cm

Table - 1
Specifications of Pendulums for Simulation
Rotational
Motor
frictional
Gear ratio (Kg)
constant (Kt)
coefficient
(B)
0.00767
70
0.004

Efficiency
of gear
(g)

Efficiency
of motor
(m)

Armature
resistance
(Rm)

90%

69%

2.6

From the above specifications, state model of RIP system in controllable canonical form is obtained as:

[
][ ] [ ]
[ ] [
][ ] [ ]

The location of closed loop poles from state model is obtained at

. One pole is lying on RH of
s-plane which means system is unstable.

V. DESIGN OF STATE FEEDBACK CONTROLLER

A. General State Feedback Controller:
The above system with state feedback gain matrix K=[K1 K2 K3] can be represented as:

Fig. 3: Representation of the System with State Feedback Controller

The overall transfer matrix from Fig 3 is obtained as:

For the system under study, :

[ ]
[
[ ][

][

[ ]

]
] [ ]

So
The transfer function including M for the above system is
B. Kharitnov Polynomial:
In the project simulation, we employ the method of Kharitonovs Theorem to determine the region forstabilizing controllerparameters (of polytypic polynomials). The main advantage of using this method is that it can be adopted or not only SISO, but
also for MIMO system of any order; even if having perturbations in them. The Generalized Kharitonov Theorem given here
provides a constructive solution to this problem by reducing it to the Hurwitz stability of a prescribed set of extremal line
segments. The number of line segments in this test set is independent of the dimension of theparameter space. This test set has
many important extremal properties that are useful in control systems.
The characteristic polynomial of the above system is given by:
(
)
( )
The most important result in this area is a theorem due to V.L. Kharitnov . In this result, the polynomial family considered is a
collection of polynomials with the following specific form:
( )
Thus, each coefficient of the polynomial is contained within a given interval. Such a polynomial family is referred to as an
interval
polynomial. Kharitonov's theorem gives a necessary and sufficient condition for the robust stability of such an interval
polynomial in terms of the following four polynomials:

All rights reserved by www.ijste.org

219

Design of Kharitnov Polynomial based State Feedback Controller for Rotary Inverted Pendulum
(IJSTE/ Volume 1 / Issue 11 / 035)

( )
( )

( )

( )
The interval polynomial is robustly stable if and only if these four polynomials are Hurwitz stable.
C. Design of State Feedback Controller by Kharitnov Polynomial:
The characteristic polynomial with state feedback controller is obtained as
( )
(
)
(
)
(
The range of
[
] for which the above polynomial becomes stable is

VI. STABILITY ANALYSIS WITH CONTROLLER

Nyquist plots for the system with statefeedback controller via Kharitnov polynomial technique are shown below:

Fig. 4: Nyquist plot for K1=7.5, K2= 38, K3= 225.5

Fig. 5: Nyquist plot for K1=-1.2, K2= 24, K3=225.5

From Fig 4 for the gain values specified, system is stable. From Fig5 for the given gain values, system is unstable.

All rights reserved by www.ijste.org

220

Design of Kharitnov Polynomial based State Feedback Controller for Rotary Inverted Pendulum
(IJSTE/ Volume 1 / Issue 11 / 035)

Fig. 6: Nyquist plot for K1=-10, K2= 100, K3=225.5

Fig. 7: Nyquist plot for K1=10, K2= 50, K3=240

From Fig 6 From Fig 4 for the gain values specified, system is unstable. From Fig7 for the given gain values ,system is stable.

VII.

CONCLUSION AND FUTURE SCOPE

In the research being carried out the system is taken as rotary inverted pendulum. It is a highly nonlinear and unstable system.
Linearization is done with the help of state space equations. From the linearized model, the transfer function of the system with
pendulum angle as output and input voltage to the motor is taken. Then from the characteristic polynomial that is the Kharitnov
polynomial is formed with state feedback controller transfer function. By giving different ranges for state feedback gain matrix,
the range which stabilizes the system is found ie all closed loop poles lies on the left side of the s plane in that range.
The future scope of the project are i) aircraft control- as a counter balancing system in the case of an aircraft ii) ship control-the
centre of gravity remains same for the research proposed. So if there are different weights on both sides of the ship, then also
ship can be stabilized on its centre of gravity by embedding this system and iii) automobile control.- During the event of different
weights on both sides of the automobile, on curve of a road the automobile can stabilized on its centre of gravity

REFERENCES
[1]
[2]
[3]

Robust swing up control for a double pendulum, Scott A. Bortoff, University of Toronto
Classical dual inverted pendulum control, Kent H Lundberg &James K Roberge, presented at 2003 IEEE conference on decision and control.
Stability Analysis and Design of PI ControllerUsing Kharitnov Polynomial for Rotary Inverted Pendulum, Jim George, Bipin Krishana, Shreesha C, V I
George, Mukund Kumar Menon, Sensors& Transducers, Vol. 138, Issue 3, March 2012,pp:104-113.

All rights reserved by www.ijste.org

221

Design of Kharitnov Polynomial based State Feedback Controller for Rotary Inverted Pendulum
(IJSTE/ Volume 1 / Issue 11 / 035)
Determination of the region of the stabilizing controller parameters for polytopic polynomials, Mukund Kumar Menon,I Thirunavakkarasu, V I George,
Sensors and Transducers Journal,Vol.119, Issue8, August 2010,pp.174-181.
[5] Minimum time swing up and stabilization of rotary inverted pendulum using pulse step control, P.Melba Mary and S.Marimuthu, Iranian Journal of fuzzy
systems Vol. 6, No. 3(2009) pp.1-15.
[6] Design,Build and Control of a Single/Double Rotational Inverted Pendulum, Final report, James Driver and Dylan Thorpe, School of Mechanical
Engineering, University of Adelaide, Australia, October 2004.
[7] Sliding mode control of suspended pendulum, Hakan Kizmaz & Sadettin Aksoy, Modern Electric power system conference 2010.
[8] Stabilising controller design for rotary inverted pendulum Bipin Krishna, I T Arasu, Dr. V I George, National conference on communication networks
and sensor technology 2011 proceedings page no: 285-288
[9] Design and Simulation of Dead Beat Model and Minimum-Order State Observer for Rotary Inverted Pendulum Proceedings of International Conference
on Science & Technology:Applications in Industry & Education (2010) (ICSTIE-2010), Universiti Teknology MARA Pulau Pinang, Malaysia, Dec-2010
[10] K Ogatta, Modern Control Theory, PHI publications, 2010
[11] ni.com/LabVIEW/ control design toolbox
[4]

All rights reserved by www.ijste.org

222