View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

provided by UTHM Institutional Repository

DEVELOPMENT OF SELF BALANCING PLATFORM

ON
MOBILE ROBOT USING PID CONTROLLER

MAZITA BINTI MAT ALI

A project report submitted in partial

fulfillment of the requirement for the award of the
Degree of Master of Electrical Engineering

Faculty of Electrical and Electronic Engineering

Universiti Tun Hussein Onn Malaysia

JUNE, 2013
 vii

ABSTRACT

This thesis presents a development self-balancing platform mobile robot using PID
controller. The platform has been designed using mobile robot kits including IMU
and two servos, and controlled by an open source microcontroller with PID. An
Arduino microcontroller, hobby grade servos, and a two-degree of freedom (axis)
accelerometer and gyroscope have been used to create the controlled platform. The
controller has been designed to maintain the platform at an initially selected angle
when the support structure orientation changes. By using Matlab, the value of PID
parameters i.e , Kp, Ki and Kd have been obtained and applied to the arduino. The
software has been written with logic to convert the digital data from the
accelerometer to an acceleration magnitude vector. The magnitude is then compared
to a predetermined mathematical function to infer the angle of tilt of the platform.
The angle of tilt is then converted to angle of rotation for the servos to act on.
Experiment shows that the platform performed as expected. Overall, the platform
design is validated based on the positional accuracy of the platform given the
relatively low quality components used to create it.
 viii

ABSTRAK

Tesis ini membentangkan mengenai pembangunan mobil robot yang mempunyai

platform keseimbangan diri menggunakan pengawal PID . Platform ini telah direka
dengan menggunakan kit robot mudah alih termasuk IMU dan dua servo serta
dikawal oleh mikro pengawal sumber terbuka dengan PID. Mikro pengawal Arduino,
servo gred hobi dan dua darjah kebebasan (paksi) pecutan dan giroskop telah
digunakan untuk mewujudkan satu platform yang terkawal. Pengawal ini telah direka
untuk mengekalkan platform pada sokongan pilihan sudut asalnya apabila orientasi
struktur berubah. Dengan menggunakan Matlab, nilai parameter PID iaitu, Kp, Ki
dan Kd telah diperolehi dan digunakan untuk Arduino . Perisian telah ditulis dengan
logik untuk menukar data digital dari pecutan ke vektor magnitud pecutan. Magnitud
ini kemudiannya dibandingkan dengan fungsi persamaan matematik yang telah
ditetapkan untuk membuat kesimpulan sudut kecondongan platform. Sudut
kecondongan kemudiannya ditukar kepada sudut putaran bagi servo untuk bertindak.
Eksperimen menunjukkan bahawa platform melaksanakan tugas seperti yang
diharapkan. Secara keseluruhan, reka bentuk platform disahkan berdasarkan
ketepatan kedudukan platform memandangkan komponen kualiti yang relatif rendah
diguna untuk menghasilkannya.
 ix

TABLE OF CONTENT

THESIS STATUS CONFIRMATION i

SUPERVISOR’S CONFIRMATION ii

DEDICATION iii

ACKNOWLEDGMENT iv

ABSTRACT vii

ABSTRAK viii

TABLE OF CONTENTS ix

LIST OF TABLES xiv

LIST OF FIGURES xv

LIST OF SYMBOLS xvii

LIST OF ABBREVIATIONS xviii

LIST OF APPENDIX xix

 x

I INTRODUCTION

1.1 Project Background 1

1.2 Problem Statement 2

1.3 Problem Objectives 3

1.4 Project Scopes 3

1.5 Organization of Report 4

II LITERATURE REVIEW

2.1 Previous Case Study 5

2.2 PID Controller 7

2.3 IMU Implementation 7

2.3.1 Gyroscope 7
2.3.2 Accelerometer 7

2.4 Mathematical Modelling 8

2.5 The Controlled platform 8

2.6 Conclusion 10
 xi

III RESEARCH METHODOLOGY

3.1 Project Methodology 11

3.2 Project Activities 12

3.3 The Hardware 13

3.3.1 Sensor IMU Board 15

3.3.2 Mobile Robot 19
3.3.3 Board Arduino Uno 20
3.3.4 DC Servo Motor 22
3.3.5 DC Motor and its Driver 25

3.4 PID Controller Design 26

3.4.1 Characteristic Of The PID Controller 28
3.4.2 Implementation of a PID controller 28
3.4.3 Tuning the PID controller 29

3.5 Model of a R/C servo Motor 30

3.6 Complementary Filter 31

3.7 Conclusion 33

IV RESULT AND ANALYSIS

4.1 Robot Design 34

4.1.1 The Experiment Result 37

 xii

4.2 Microcontroller Programming 39

4.3 Sensors Integration 42

4.3.1 Gyrometer 42

4.3.2 Accelerometer 43

4.4 Filter Design 47

4.4.1 Integration 47

4.4.2 Low-Pass Filter 47

4.5 Simulation Result 48

4.5.1 Continuous to Discrete Conversion 49

4.5.2 Pitch Axis Control 51

4.5.2.1 PID control in Pitch Axis 51

4.5.2.2 Tuning the gains in Pitch Axis 54

4.5.3 Roll Axis Control 55

4.5.3.1 PID control in Roll Axis 57

4.5.3.2 Tuning the gains in Roll Axis 58

4.6 PID Control Analysis 60

4.7 Conclusion 62
 xiii

V DISCUSSION AND CONCLUSION

5.1 Discussion 64

5.2 Conclusion 65

REFERENCES 67

APPENDIX A 70

APPENDIX B 72

APPENDIX C 76
 xiv

LIST OF TABLES

2.1 List of Platform Projects 10

3.1 Characteristic of PID Controller 28

4.1 Measurement Result For RC Servo 38

4.2 Angle Estimate by Experiment 48

4.3 Analysis of PID controller result for Roll and

pitch axis. 60
 xv

LIST OF FIGURES

3.1 Flow Chart of process Methodology PS1 11

3.2 Flow Chart of process Methodology PS2 12
3.3 Platform Schematic 13
3.4 External installation Components 14
3.5 IMU sensor 6 DOF 15
3.6 Gyro Axes (θ ) Relative to PCB Axes ( θ) 17
with Orientation Errors (θ and Φ)
3.7 Orientation of Axes of Sensitivity and Polarity 17
of Rotation +Z +X +Y
3.8 The Chassis of mobile robot 19
3.9 Differentially steered 3 wheeled Mobile Robot 20

3.10 Arduino Development board 21

3.11 Atmel Atmega 328 microcontroller 21

3.12 The graphical user interface (GUI) of Arduino Compiler 22
3.13 Servomotor circuit diagram 22
3.14 Servo Arrangement 23
3.15 Servo Body Brackets 24
3.16 Servo "C" Brackets 24
3.17 An Image of the DC motor driver. 25
3.18 Block diagram of PID closed loop 26
3.19 Closed Loop control for X-axis output. 27
3.20 Closed Loop control for Y-axis output. 27
 xvi

3.21 Block diagram of an RC Servo motor 30

3.22 Electro-Mechanical scheme of a DC motor 30
3.23 Complementary Filter in block diagram form. 32
4.1 Mobile robot with a platform is in a stationary 34
4.2 Mobile robot with a platform on upward ramp. 34
4.3 Mobile robot with a platform on downward ramp 36
4.4 Design Of Self Balancing Platform On Mobile Robot 37
4.5 Degree of servo Vs Time 39
4.6 Voltage Vs Time 39
4.7 Sketch Flow Chart 41
4.8 Accelerometer Gravity Measurement 45
4.9 Stairstep open loop response 49
4.10 Stairstep close loop response 50
4.11 Output Response of step Input Kp=5 51
4.12 Output Response for Kp=5 with 5% step disturbance 52
4.13 Output Response of step input Kp=5 Ki=0.01 53
4.14 Output Response for Kp=5 Ki=0.01 with 5% step disturbance 53
4.15 Output Response of step input Kp=5 Ki=0.01 Kd=0.01 54
4.16 Output Response for Kp=5 Ki=0.01 Kd=0.01
with 5% step disturbance 55
4.17 Output Response of step Input Kp=2 56
4.18 Output Response for Kp=2 with 5% step disturbance 56
4.19 Output Response of step Input for Kp=2 Ki=0.1 57
4.20 Output Response for Kp=2 Ki=0.1 with 5% step disturbance 58
4.21 Output Response of step input Kp=2 Ki=0.1 Kd=0.01 59

4.22 Output Response for Kp=2 Ki=0.1 Kd=0.01

with 5% step disturbance 59

4.23 Closed loop for pitch (axes-Y) 61

4.24 Closed loop for roll (axes-X) 61

 xvii

LIST OF SYMBOL

G - Gravity
θ - Angle
t - Time
J - Moment of Inertia
e(t) - Back Electromotive Force
τ - Torque
I - Current
R - Electric Resistance
L - Electric inductance
Ke - Back electromotive Force Co-efficient
B - Damping Ratio Of The Mechanical System
V - Voltage
ms - Millisecond
 xviii

LIST OF ABBREVIATIONS

Xmeas - Measurement of axis–X

Ymeas - Measurement of axis-Y
IMU - Inertial Measurement Unit
Vref - ADC reference voltage
Vzero rate - Zero rate Voltage
 xix

LIST OF APPENDICES

APPENDIX TITLE

A Gantt chart for PSM 1 and PSM 2 69

B Project Coding Matlab 71
C Project Coding ARDUINO 75
 1

CHAPTER 1

INTRODUCTION

1.1 Project Background

Designing a mobile robot with special capabilities has become a trend these days for a
variety of universal human consumption. It also fits well with the needs and nature of
the human lifestyle. Different forms and uses, mobile robots have been designed and
are now in the market worldwide.
A mobile robot comprises of three main parts including sensors, logical
processing unit and actuator. In this project, a robot that can maintain an upright and
balanced position on a platform is designed and developed. The robot consists of
Inertial Measurement Units (IMU) sensors, microprocessor and motors. The design is
designed with Matlab and the resulting parameters are used and burned into Arduino
UNO controller. The main purpose of the controller is to fuse the wheel encoder,
gyroscope and accelerometer sensors to estimate the attitude of the platform and then
to use this information to drive there action wheel in the direction to maintain an
upright and balanced position platform.
If the platform system itself is not balanced, which means it keeps falling off
away from the vertical axis, then a gyro chip is needed to provide the angle position of
the inverted pendulum or robot base and input into the controller, which the program
itself is a balancing algorithm. The PID controller will then provide a type of feedback
signal through PWM control to turn the motor servo clockwise or anticlockwise, thus
balancing the platform. These two measurements are summed and fed-back to the
actuator which produces the counter torque required to balance the platform robot.
 2

In this project, the PID will be used because it is relatively easy to implement
yet practical. Besides that, PID controller has only three adjustable parameters that can
be determined from several techniques. Previous research has shown that PID
controller has shown good results in terms of response time and accuracy when the
parameters i.e, Kp , Ki and Kd, are properly tuned.

1.2 Problem Statements

Control systems are often designed to improve stability, speed of response, steady-
state error, or prevent oscillations. Many researchers wants to produce a mathematical
equation that is able to determine the position of a very accurate motor position, thus
the steady state error should be zero. DC motor systems have played an important role
in the improvement and development of the industrial revolution. Therefore, the
development of a more efficient control strategy that can be used for the control of a
DC servomotor system and a well defined mathematical model that can be used for off
line simulation are essential for this type of systems. Servomotor systems are known to
have nonlinear parameters and dynamic factors, so to make the systems easy to
control, conventional control methods such as PID controllers are very convenient.
Also, the dynamics of the servomotor and outside factors add more complexity to the
analysis of the system, for example when the load attached to the control system
changes.
Due to these parameters and factors, this study will apply the PID controller to
make the steady-state error, due to continuous disturbance, to be zero. Accordingly
this project will review the principles of PID that is used to control the servo
movement that depends on the angle captured by the IMU. This project uses the PID
to compensate the robot body inclination to stabilize the platform. Among other
performance requirements are to reach the final position of the motor position very
quickly without excessive overshoot. In this case, focusing on systems that have a
finish time of 10 ms and the overshoot is smaller than 25%.
.
 3

1.3 Project Objectives

The aim of this project is to implement PID controller to a mobile robot to maintain its
flatness on a moving platform. The objectives of this project are as follows:
a) To design and develop a mobile robot and a flat platform.
b) To design a PID controller to maintain the robot flatness
c) To simulate the controller using Matlab and analyse its performance.
d) To integrate the controller into the mobile robot.

1.4 Project Scopes

The scopes of study are as follows:

a) Using mobile platform kits available in the market.
b) Using sensor fusion to measure the tilts in the X-axis
and Y-axis.
c) Using PID as the flatness controller.
d) Using arduino Uno as the mainboard of the mobile robot.

1.5 Organization of Report

As an overview, the structure of this report is organized as follows :

Chapter 1 describes a general introduction of the project, problem statement project
aims and project scope.

Chapter 2 provides details literature review that includes an introduction to some

basic concepts and a survey of existing works in the areas of developing an algorithm
for solving PID controller to maintain flatness platform. This chapter explains in detail
all the researches, studies, theories and gathering that have been make throughout the
project.
 4

Chapter 3 discusses the methodology of the project which provides a detailed

description of the design to develop a mobile robot using PID controller to maintain
the platform flatness. It also discusses about the hardware which has been used in the
project.

Chapter 4 discusses about the result and analysis. It also includes the design of a
mobile robot with a flat platform.

Chapter 5 concludes the project and gives suggestions for future work.
 5

CHAPTER 2

LITERATURE REVIEW

2.1 Previous Case Study

Conducting initial review research is very critical in understanding self

balancing platform control techniques. The review of research about related
literature conducted in this project summarizes some of topics related to the
techniques used for the balancing of platform based on Dc motor position.
Comparisons between the present project and the related topics of existing
information will also be discussed. The methodologies and the techniques used
by other researchers around the globe on the balancing platform topic will also
be reviewed.
Meena et al. (2011) proposed a design for a servo motor controller in
discrete-time system to obtain the transfer function of the PID controller design.
MATLAB / Simulink has been used to confirm the effectiveness of this new
design method, which provides a simple and powerful way to design a speed
controller for servo motor.
It also extracted a DC servo motor mathematical model and equations
and there were three different motion controllers that were designed and
simulated to control the velocity of the motor.
Popescu et al. (2011) did a comparison between PID and Fuzzy
controllers used in mobile robot control. There is a significant problem for fuzzy
controllers in which computing time is longer than the PID because a lot of
complex operations such as requiring fuzzification, inference, and
defuzzification.
 6

Masakazu et al. (2005) proposed a tuning method for PID controller that
considers changes in system characteristics. It is about the concept of using the
optimization of PID controller tuning, depending on the obstacles on the control
input derivatives and considering model uncertainties caused by changes in the
system dynamics. Partial model matching method was used to evaluate
performance and control while the reference referred to interference and
repression compared to the tracking properties.
Arpit et al. (2012) proposed a performance comparison of PID and Fuzzy
logic controller using different defuzzification techniques for positioning control
of dc motors. The result of the fine-tuned PID controller gives relatively less
overshoot and settling time with no steady state error. The fuzzy logic controller
with different defuzzification techniques gives zero % overshoot and lesser
settling time.
In a paper titled ‘Attitude Estimation Using Low Cost Accelerometer and
Gyroscope’ written by Young Soo Suh (2003), it shows two different sensors
which are the accelerometer and gyroscope that exhibit poor results when used
separately to determine the attitude which is referred as the pitch angle or roll
angle. However, the gyroscope can combine with accelerometer to determine the
pitch or roll angle with much better result with the use of Kalman filter.
Tomislav et al (2012) proposed self-balancing mobile robot tilter. It
provides a summary of work done in the field of electronic, mechanical design,
software design, system characterization and control theory. Robotic system
model and simulation results of various control methods required for the
stabilization of the system were studied. Dynamic effects become increasingly
important in assessing performance limits in robotic. The processes where the
project was carried out including design and production of certain parts of the
integration section, electronic, mechanical and software.
Hany Ferdinando et al. (2001) proposed a paper titled Developing
Mathematical Model of DC Servo Motor Using Bond Graph. Bond graph can
serve as a tool in the modeling of some plants. The plant was DC servo motor
with all parameter is unity. Simulink and 20-Sim was used for the simulation
procedure.
 7

The use of plant transfer function had several performances, such as using a step
response and unit impulse function, the parameters change, how to control the
torque and speed using the input voltage and current.

2.2 PID Controller

This project concerns the development of a mobile robot with a platform, which
can be levelled using PID controller. The main objective is to control the flatness
of the platform efficiently with a low cost hardware without limiting the strength
and performance of the whole system. There are various stages that have been
used to stabilize the platform such as modelling the system, obtaining the data
from sensors and determining how the control algorithms will be implemented.
V.J. Van Doren (2009) suggested a two wheeled robot to perform the balancing
and control of mobile robots. In this project the Proportional, Integral,
Derivative (PID) has been implemented to control the flatness of a mobile robot
platform. PID has proven to be popular among the control engineering
community.
As stated by the author of article Vance J. VanDoren (2009), “For more
than 60 years after the introduction of Proportional-Integral-Derivative
controllers, remain the workhorse of industrial process control”.
PID controller will be extensively discussed in Chapter 3.

2.3 IMU Implementation

2.3.1 Gyroscope

A gyroscope is a device that measures angular rate around an axis. Tilt angle can
be obtained by integrating angular rate over sampled time. The gyroscope
modifies a servo control signal by an amount proportional to its measure of
angular velocity. An estimate of angular displacement is obtained by integrating
the velocity signal over time. The Inclinometer outputs an analogue signal,
 8

proportional to the angular displacement of the sensor Braun,T.,Sutherland ,and

Alistair (2002). More about gyroscope can be found in Chapter 3.

2.3.2 Accelerometer

As stated by A. Warnasch, and A. Killen (2002), the tool that measures the
inertial force in the opposite direction of free fall acceleration vector in terms of
g-force is acceleration. So, gravity acceleration shows 0g power during free fall
down constantly at 1g. It will be supported by the ground with a force equivalent
of 1g, when the accelerometer is at rest on the surface of the earth. Because of
that, it will show a constant downward force perpendicular to the ground 1g rest.
To obtain the tilt angle using the Pythagorean Theorem can also use these
features. Readers are referred to Chapter 3 for more on accelerometer.

2.4 Mathematical Modelling

Modelling is the process of identifying the principal physical dynamic effects to

be considered in analysing a system, writing the differential and algebraic
equations from the conservation laws and property laws of the relevant
discipline, and reducing the equations to a convenient differential equation
model (Robert, 1999). In order to develop the control system, mathematical
model is established to predict the behaviour before applied into real system.
Actually, the dynamics refer to a situation which is varying with time (Ernest,
1972). The dynamic performance of a balancing robot depends on the efficiency
of the control algorithms and the dynamic model of the system. Mathematical
modelling of DC motor will be discussed in Chapter 3.

2.5 The Controlled platform.

The controlled platform is a popular university level project, where many teams
from various colleges and institutions as well as independent parties (tinkerers)
complete designs for credit. A web search of the terms “controlled platform” or
 9

“self-leveling platform” yields a number of different platform concepts of

varying complexity and sophistication. Table 1 below lists the qualities of each
design and the differences between them and the concept proposed by this paper.
The author selected eight completed designs for comparison. These designs
were selected using their similarity in form and function to the proposed
platform as the main criteria.
Note that some of the platforms do not have a complete design
description or parts list. Those will be compared using the pictorial or video
evidence available via the references.

Table 2.1 - List of Platform Projects

Platform Budget Electronics Hardware Build Team Angle

Description Sophistication Control
[Reference]
Auto-level Mid- Arduino Mid-level St. Mary’s none
[19] level Digital Servos Two axis U. single
person team

Automotive Mid- PIC18F452 Mid-level Devry New none

self-leveling level micro- ~Three axis Brunswick
[20] controller ~3 person
team
Self-leveling Low- Arduino Low-level Independent none
Surface level Single axis party
[21]

Self-leveling High- ezDSP High-level Berkeley none

platform level Three axis 4 person
[22] team

Automotive Mid- unknown Mid-level Unknown none

self-leveling level Two axis Institution
[23]
self- leveling Mid- dsPIC Mid-level Unknown none
platform level Single axis Institution
[24]

Stewart High- unknown High-level U of Adger, likely

Platform level PC Control Six axis Norway
[25]
Self-leveling High- PC Control High-level Unknown, none
platform level via Hitachi Two axis Germany
[26] SH2
 10

2.6 Conclusion

In this chapter, an overview of the previous research on robot control has been
done, including the IMU sensors and mathematical modelling. In addition, a
brief discussion on PID controller has been put forward. It is found that the
research on self-levelling platform is a popular topic. As such, a number of
research on the topic have been performed and published their findings. This
chapter has summarised some of them in Table 2.1.
 11

CHAPTER 3

RESEARCH METHODOLOGY

3.1 Project Methodology

This project is done in three phases. The first phase is to understand and design
the mobile robot as well as to implement the theories in the real hardware. The
second phase is to understand the PID controller and its characteristic and to
design and implement the controller into the robot. The last phase is to analyse
the controller performance using M-file in Matlab and compare it with the
response of the real hardware.
Figure 3.1 shows a general procedure of the project implementation
during PS1 while Figure 3.2 illustrates the procedure during PS2.

Start PS1

Research topic selection

Decide the project’s tittle , objective and scope

Literature review and methodology

Design a mobile robot with a flat platform.

PS2

Figure 3.1 : Flow Chart of process Methodology PS1

 12

PS2

Design controller using PID hand on the

derived mathematical model

Analysis the suitable parameter and controller

structure using matlab/Simulink.

Test

Implement the controller to real hardware.

Evaluate the performance of the designed

PID controller

End

Figure 3.2 : Flow Chart of process Methodology PS2

3.2 Project Activities

The process of executing this project is quite challenging because it involves

several phases which include understanding the underlying theories, designing
the robot as well as the PID controller. To make the process systematic, a Gantt
Chart has been formulated. Gantt Chart is very important in order to complete
this project on time. It starts with discussing with supervisors on the topic for
the project, objectives, scopes and related things. Study on the literature is also
planned in the Gantt Chart. Before proceed the to the PID controller design
phase, it is very important to learn how to use the Simulink Matlab, which is
also stated in the chart. Implementation and works of the project are
summarized into the Gantt chart, as shown in appendix A.
 13

3.3 The Hardware

The self-correcting platform consists of two platforms (Figure ), the “top”

platform (smaller wood piece), which is autonomously controlled, based on
initial user input, which is where the IMU sensor is installed and the angle of
tilt is measured.

Figure 3.3 : Platform Schematic

An Arduino Uno board, ADXL325 accelerometer, ITG 3205

Gyrometer , two MMG 995 servos, and a Li Po battery control the upper
platform tilt angle. The following sections describe the mechanics, electronics,
and programming aspects of the platform. The Arduino board connections to
the IMU sensor and servos are illustrated in
Figure 3.4. Normally, the Arduino board and other accessories can be
powered from the universal serial bus (USB) connection used to program the
board.
 14

DC Motor
Driver
Left
ire

Stacked

Right
IMU
Sensor

Stabilized
Platform

Stacked

Arduino
UNO
Power
Supply

Voltage
Regulator

Battery

Figure 3.4 : External installation Components

However, the servos used for this board require more power than the Arduino
board can supply from the 5V power pin. Because of this, the servos are
powered using an external power source, a 9V battery.
According to the MMG 995 servo data , the servos are rated to 6V max,
so the 9 volts from the battery are converted to 5 volts via a 5V, 1A power
regulator. The regulator has enough power capacity to drive both servos at
once with low load.
 15

The servomotor used to manually control the platform are supplied with
5V from the Arduino board. It is the reduction in voltage supply to the analog
pins which signals the board to send an angle signal to the servos. The
accelerometer is powered by the Arduino board’s on-board 3.3V power. The
ADXL325 accelerometer will send different signals through the analog pins
depending on the supply voltage.

3.3.1 Sensor IMU Board

To address the issue of measurement noises and the limitations of

measurements by either the accelerometer or gyroscope alone, we will need to
combine the readings from both the accelerometer and the gyroscope in a
meaningful way so that we could use the strengths from both sensors to obtain
a more accurate result than either measurement alone could provide.

Figure 3.5 : IMU sensor 6 DOF

Young (2003) states that the factor that contributes to the deviation of
the desired result of the gyroscope is the drift term. Since the drift increases
with time error in output data will also increase.
 16

One of the disadvantages of using accelerometer individually is that the

device is sensitive to vibration since vibration contains lot of acceleration
components. One solution that Young suggested is that a low pass filter is
required to limit the high frequency.

a) Sensor Gyroscope

Gyroscope can measure the rate at which the rotation is taking place and the
rotation angle for a given time interval is governed by:

(t)= t1t2G(t)dt (3.1)

where G(t) is the gyroscope reading with respect to the rotation direction.
When the time interval is small, the gyroscope reading can be treated as a
constant and can be approximated as a above equation:

(t) (t1)+G(t)(t2−t1)= (t1)+G(t) t... (3.2)

Gyroscope measurement is largely immune to none angular movement

and thus far less susceptible to vibrations and lateral accelerations mentioned
previously. Drifting effect will happen if the angular measurement is
cumulative, any minute error in measurements will manifest over time. Thus
gyroscope alone cannot be used to reliably measure the inclination angle either.
Orientation errors of the gyroscope mounted to the printed circuit board
can cause cross-axis sensitivity in which one gyro responds to rotation about
another axis, for example, the X-axis gyroscope responding to rotation about
the Y or Z axes. The orientation mounting errors are illustrated in Figure 3.6.
 17

Figure 3.6 : Gyro Axes (θ ) Relative to PCB Axes ( θ) with Orientation Errors
(θ and Φ)

Orientation

Figure 3.7 shows the orientation of the axes of sensitivity and the polarity of
rotation.

Figure 3.7 : Orientation of Axes of Sensitivity and Polarity of

Rotation +Z +X +Y

b) Sensor accelerometer

Accelerometers are mechanical structures containing elements that are free to

move. These moving parts can be very sensitive to mechanical stresses, much
more so than solid-state electronics. The 0 g bias or offset is an important
 18

accelerometer metric because it defines the baseline for measuring

acceleration. Additional stresses can be applied during assembly of a system
containing an accelerometer. These stresses can come from, but are not limited
to, component soldering, board stress during mounting, and application of any
compounds on or over the component. If calibration is deemed necessary, it is
recommended that calibration be performed after system assembly to
compensate for these effects.
For the need to get a good measurement of the current inclination angle
in order to control the platform’s movements. Inclination Angle Calculation
can can be calculate as:

=tan−1AyAx=sin−1Ax A2x+A2y=sin−1gAx (3.3)

In the equations above, Ax is the accelerometer reading along its x axis

and Ay is the accelerometer reading on its y axis. When the robot is stationary,
g is the gravitational constant (which translates into the accelerometer reading
on the y axis when the accelerometer is lying flat). In this studies, are interested
in getting a calculation in which the tilt angle is small because our goal is to
ensure that the deviation from equilibrium (usually equilibrium is reached
when the inclination angle close to 0 depending on the weight distribution of
the robot) is as small as possible. So that can further simply the above equation
for small inclination angles:
 sin( ) = gAx... (3.4)

By only measuring the x axis reading of the accelerometer, we can get a

rough estimate of the inclination angle. Of course, this is under the assumption
that the robot is standing still. At the same time, the y axis reading will be
slightly less than the reading in standstill condition (Ay). As a result, the
combined vector will deviate from g. But when the accelerometer is placed
near the center of gravity of the platform, the acceleration along the x axis is
small in near-equilibrium conditions. Accelerometer tends to be very sensitive
to the accelerations introduced due to movement or vibration and thus the
sensor readings will contain some level of noise, which cannot be removed
easily.
 19

The values measured for X0g and Y0g correspond to the x- and y-axis
offset, and compensation is done by subtracting those values from the output of
the accelerometer to obtain the actual acceleration:

XACTUAL = XMEAS − X0g

YACTUAL = YMEAS − Y0g

The ADXL345 can automatically compensate the output for offset by

using the offset registers (Register 0x1E, Register 0x1F, and Register 0x20).
These registers contain an 8-bit, twos complement value that is automatically
added to all measured acceleration values, and the result is then placed into the
DATA registers.
Because the value placed in an offset register is additive, a negative
value is placed into the register to eliminate a positive offset and vice versa for
a negative offset. The register has a scale factor of 15.6 mg/LSB and is
independent of the selected g-range.

3.3.2 Mobile Robot

Mobile robots that are used in this project is ready made. It features two gear
motors with 65mm wheels and a rear caster. The chassis plates are cut from
acrylic with a wide variety of mounting holes for sensors, controllers, power,
etc. Simply bolt the two pre-cut platforms together, attach the motors and
caster and robotics controller. It includes all of the parts needed to assemble the
chassis as well as a 4xAA battery holder with barrel jack termination.

Figure 3.8 : The Chassis of mobile robot

 20

a) 3 Wheels Mobile Robots

The 3-wheel mobile robot has two rear wheels and an independently rotating
front wheel. The robot can change direction by varying the relative rate of
rotation of two separately driven rear wheels. If both wheels are driven in the
same direction and speed, the robot will go straight.
The rotation center can fall anywhere in the line joining two wheels.
The center of gravity in this type of robot has been put in the triangle formed
by the wheel. If too heavy mass attached to the free side of the spinning wheel,
the robot will tip over.

Figure 3.9 : Differentially steered 3 wheeled Mobile Robot Source [28]

3.3.3 Board Arduino Uno

The main controller for this project is Arduino Uno which uses Atmel Atmega
328 microcontroller as it combines 32KB ISP flash memory with read-while-
write capabilities, 1KB EEPROM, 2KB SRAM, 23 general purpose I/O lines,
32 general purpose working registers, three flexible timer/counters with
compare modes, internal and external interrupts, serial programmable USART,
a byte-oriented 2-wire serial interface, SPI serial port, 6-channel 10-bit A/D
converter (8-channels in TQFP and QFN/MLF packages), programmable
watchdog timer with internal oscillator, and five software selectable power
saving modes.
 21

The device operates between 1.8-5.5 volts. The overview of the device
as shown in figure 3.9 and figure 3.10. (Arduino.cc, 2012).

Figure 3.10: Arduino Development board

Figure 3.11: Atmel Atmega 328 microcontroller

Arduino is an open-source electronics prototyping platform based on

flexible, easy-to-use hardware and software. It's intended for artists, designers,
hobbyists, and anyone interested in creating interactive objects or
environments.
The open-source Arduino environment as shown in Figure 3.10 makes
it easy to write code and upload it to the I/O board. It runs on Windows, Mac
OS X, and Linux. The environment is written in Java and based on Processing,
avr-gcc, and other open source software.
 22

Figure 3.12: The graphical user interface (GUI) of Arduino Compiler

3.3.4 DC ServoMotor

DC servomotors are one of the main components of automatic systems; any

automatic system should have an actuator module that makes the system to
actually perform its function. The most common actuator used to perform this
task is the R/C servomotor. A servomotor system consists of different
mechanical and electrical components, the different components are integrated
together to perform the function of the servomotor. Figure 3.13 above shows a
typical model of a servomotor system (Nise, 2008).

Figure 3.13: . Servomotor circuit diagram.

 23

The servos supporting and controlling the top platform are arranged in
order to control the pitch and roll. Note that pitch and roll are only subjective
directions and are used to describe the motion. Since the platform has no
“front” or “back” the terms pitch and roll are meaningless descriptions of the
rotations about the X and Y axis. A close up view of the servo arrangement
and axis’ is illustrated in Figure 3.14.

Figure 3.14 : Servo Arrangement

This servo arrangement is a more efficient use of space and limits the
amount of slop compared to linkages and hinges as seen in other designs. The
Y axis servo is controlled by the Y axis data stream from the accelerometer and
vice versa for the X axis. Note that the X axis servo motion is in-plane with the
photo while the Y axis would rotate out of plane.
The servos are installed into aluminum brackets illustrated in
Figure 3.15. These brackets allow the servo to be installed onto the
bottom and top platforms. The servo “C” brackets (
Figure 3.16) are then used to attach the servo bodies, via the body
brackets, to themselves and the platforms. The “C” brackets attach to the body
and to the rotating servo horn.
The combination of these two brackets allows the platforms to rotate
independently in two axis.
 24

Figure 3.15 : Servo Body Brackets [19]

Figure 3.16 : Servo "C" Brackets [19]

A dual axis accelerometer was the chosen to measure the tilt of the
lower platform in two axes. The data is read from two different channels and
processed in the same manner regardless of direction. The two analog data
streams provide the Arduino logic with the necessary information to maintain
the top board level to the chosen plane.
 67

REFERENCES

[1] R.S.Meena,Vikas Kumawat(2011) .Controller Design For Servo Motor

Using MATLAB .In Proceeding of National Conferences on Advances &
Research in Electrical System Technology ( AREST 2011).

[2] Popescu, Cristina; Paraschiv, Nicolae; Cangea, Otilia.(2011) Comparison

Between Pid And Fuzzy Controllers Used In Mobile Robot Control
annals of daaam & proceedings;jan2011, p223

[3] Kotaki, Masakazu; Yamakawa, Yuji; Yamazaki, Takanori; Kamimura, A

Tuning Method for PID Controller That Considers Changes in System
Characteristics. ASHRAE Transactions. 2005, Vol. 111 Issue 2, p13-22. 10p.

[4] Arpit Goel., et al. (2012) Performance Comparison Of PID And Fuzzy Logic
Controller Using Different Defuzzification Techniques For Positioning
Control Of DC Motors. Journal of Information Systems and Communication
ISSN: 0976-8742 & E-ISSN: 0976-8750, Volume 3, Issue 1, pp.-235-238.

[5] Young Soo Suh.(2003) .Attitude Estimation Using Low Cost Accelerometer
And Gyroscope.Proceedings KORUS 2003. The 7th Korea-Russia
International Symposium on Volume: 2 Page(s): 423 - 427 vol.2

[6] Andrea Demetlika, Tomislav Tomašić, Mladen Crneković (2012)

Selfbalancing Mobile Robot Tilter.FAMENA issue 3, volume 36, Zagreb

[7] Hany Ferdinando, Heri Saptono Warpindyasmoro, Stanley Kardinal Jusuf.

(2001) Developing Mathematical Model of DC Servo Motor Using Bond
Graph 1st Kentingan Physics Forum ,Surakata

[8] V.J. VanDoren, “PID: Still The One,” Control Engineering, October 2003.
Retrieved

[9] D. Simon, Kalman Filtering With State Constraints: A Survey Of Linear And
Nonlinear Algorithms, IET Proceediings in Control Theory & Applications,
vol. 4, no. 8, pp. 1303-1318, 2010.

[10] Braun,T.,Sutherland ,and Alistair (2002).An Experimental Platform for

Researching Robot Balance.Department of Electrical & Electronic
Engineering .University of Western Australia.
 68

[11] A. Warnasch, and A. Killen, Low cost, high G, micro electro-mechanical

systems (MEMS), inertial measurements unit (IMU) program, IEEE
Position Location and Navigation Symposium, 2002, pp. 299-305.

[12] Robert L.Woods(1999).Modeling And Simulation Of Dynamic

Systems.New Jersey: Prentice Hall.

[13] Ernest O. Doebelin.(1972).System Dynamics.Modeling And

Response.USA:Charles E. Merril Publishing Company.

[14] Makableh, Y. (2011). Efficient Control Of DC Servomotor Systems

Using Backpropagation Neural Networks.

[15] Nise, Norman S.. Control Systems Engineering. Fifth edition 2008.

[16] MATLAB/Simulink User’s Guide, 1998, The Math Works Inc, Natick,
MA,

[17] Retrieved on Jan 15 ,2013 ,from

http://www.princeton.edu/~mae412/TEXT/NTRAK2002/292-302.pdf

[18] Arduino.cc. (2012), Arduino Uno Rev 3. Retrieved on Jan 15, 2013,
http://arduino.cc/en/Main/ArduinoBoardUno

[19] Auto-Leveling Platform – St. Mary’s University

http://engineering.stmarytx.edu/~nechon/

[20] Devry New Brunswick

http://www.youtube.com/watch?v=f9ALAvE3gBQ

[21] Self -leveling surface with arduino

http://www.youtube.com/watch?v=cTUBDagKdbA&feature=related

[22] Self-Leveling Platform

http://www.me.berkeley.edu/ME102/Past_Proj/f08/group_09/
intro_objective.html

[23] Self-leveling platform

http://www.youtube.com/watch?v=TiTRUwU7kRs

[24] ECE572SelfLevelingPlatform
http://www.youtube.com/watch?v=CuN_ZkLK0gM&NR=1

[25] Stewart Platform University of Adger, Norway

http://www.youtube.com/watch?v=WmKnnp1xTPg&NR=1

[26] Self-leveling platform for control of autonomous lawn mower

http://www.dspace.de/ftp/papers/uc/cheok.pdf
 69

[27] Proposed self-leveling controlled platform

http://www.youtube.com/watch?v=vmf1ThwrNM0

[28] Retrieved on Apr 15 ,2013 ,from

http://en.wikibooks.org/wiki/Robotics/Types_of_Robots/Wheeled

[29] Retrieved on Apr 15 ,2013 ,from

http://www.starlino.com/imu_guide.html