The Modelling of Tyre Rotation Behaviour With

THE MODELLING OF TYRE ROTATION BEHAVIOUR WITH

TYRE PRESSURE MONITORING SYSTEM

GOH YIK CHOONG

UNIVERSITI SAINS MALAYSIA

2017 THE MODELLING OF TYRE ROTATION BEHAVIOUR WITH

TYRE PRESSURE MONITORING SYSTEM

GOH YIK CHOONG

Thesis submitted for partial fulfillment of the requirement for the degree of Master of Science (Microelectronic Engineering)

August 2017 ACKNOWLEDGEMENT

First and foremost, I would like to express my deep gratitude to research supervisor Dr. Mohamed Fauzi Bin Packeer Mohamed, for his patient guidance, enthusiastic encouragement and constructive suggestion on my research planning and development. His willingness to give his time, starting from first semester before the actual research start date. In addition, his professionalism and passion that supported me throughout the Master program until completion is much appreciated.

Besides that, I would also like to thank Intel technology as well as my direct superior Mr. Lim Choon Aun for offering me the opportunity enroll in this postgraduate micro-electronic master program and his support is much appreciated. Apart from that, I wish to thank various people from USM and Usains that were involved in contributing to this Mixed-mode partnership master program and making it a success.

Last but not least, warmest thought to my lovely family members for supporting my decision in every milestone in my life. Without their continuous support, I would not have the opportunities to achieve success, especially during the challenging period to carry on my research work.

TABLE OF CONTENTS

ACKNOWLEDGEMENT ...... i

TABLE OF CONTENTS ...... ii

LIST OF TABLES ...... v

LIST OF FIGURES ...... vii

LIST OF ABBREVIATIONS ...... x

ABSTRAK ...... xii

ABSTRACT ...... xiii

CHAPTER 1 INTRODUCTION ...... 1

1.1 Background ...... 1

1.2 Problem Statements ...... 3

1.3 Research Objectives ...... 4

1.4 Research Scope ...... 4

1.5 Research Contribution ...... 5

1.6 Thesis Outline ...... 6

CHAPTER 2 LITERATURE REVIEW ...... 7

2.1 Introduction ...... 7

2.2 Existing Tyre Monitoring Technologies ...... 7

2.2.1 Direct and Indirect Measurements ...... 7

2.2.2 Survey and Comparison of Tyre Monitoring Technologies ...... 9

2.3 Type Of Tyre Construction, Improper Inflations And Tyre Depth Measurement ...... 11

2.3.1 Bias-ply and Radial Tyre Construction ...... 11

2.3.2 Tyre Failure Caused by Improper Inflation ...... 14

2.4 Wireless Technologies ...... 16

2.4.1 Overview of Technologies Survey in Wireless Communication for Vehicle ...... 16

2.4.2 Operation Mode ...... 17

2.4.3 Frequency, Data Rate and Range ...... 18

2.4.4 Power Consumption ...... 18

2.5 Review on Tyre Rotation Behaviour Model ...... 19

2.5.1 Tyre Rotation under Different Revolution ...... 19

2.5.2 Tyre Rotation under Different Inflation ...... 22

2.5.3 Data Transmission in Vehicle...... 25

2.6 Chapter Summary ...... 27

CHAPTER 3 METHODOLOGY ...... 30

3.1 Introduction ...... 30

3.1 Methodology of Project ...... 30

3.2 Block Diagram of Tyre Pressure Monitoring System ...... 31

3.2.1 Transmitter Module ...... 33

3.2.2 Receiver Module ...... 33

3.3 Flow Chart ...... 33

3.4 Selection Of Hardware ...... 37

3.4.1 Pressure Sensor ...... 37

3.4.2 Inertia Measurement Unit (IMU) ...... 38

3.4.3 Wireless technologies ...... 39

3.4.4 Bluetooth ...... 40

3.4.5 Microcontroller ...... 41

3.5 Design & Experiment Setup For Tyre Pressure Monitoring System ...... 42

3.5.1 Design of Transmitter Module inside the Tyre ...... 43

3.5.2 Transmitter and Receiver Module Circuit Design ...... 45

iii

3.5.3 Experimental Setup...... 48

3.5 Chapter Summary ...... 53

CHAPTER 4 RESULTS AND DISCUSSIONS...... 54

4.1 Overview ...... 54

4.2 Modelling Of Vehicle Rotation Behaviour and Vehicle Distance Traveled Calculation ... 54

4.2.1 Accelerometer Performance Based On Wheel Rotation with Speed Variation ...... 54

4.2.2 Accelerometer Performance Based On Wheel during Braking / Stop Condition ...... 57

4.2.3 Accelerometer Performance Based On the Wheel Rotation Counter And Distance Travel Calculation ...... 60

4.3 Monitoring On Pneumatic Pressure inside Tubeless Tyre ...... 64

4.3.1 Investigate From Atmospheric Pressure Level to Project Specification ...... 64

4.3.2 Pressure Sensor Reading Accuracy Test by Reverse Engineering ...... 69

4.4 Data Transmission Quality on Bluetooth Connection...... 72

4.5 Chapter Summary ...... 75

CHAPTER 5 CONCLUSION AND FUTURE WORKS ...... 77

5.1 Conclusion ...... 77

5.2 Future Works ...... 78

REFERENCES ...... 79

APPENDIXES ...... 1

LIST OF TABLES

Table 1.1 Statistic of General Road Accident Data in Malaysia [1]……..…………..… 1

Table 1.2 Statistic of Road Deaths by Type of Vehicle [2]……………………………... 2

Table 1.3 Vehicle Related Critical Reasons [3]…………………………………………. 3

Table 2.1 The Comparison of Tyre Monitoring Technologies Working

Principle [13, 14]………………………………………………………………………….. 10

Table 2.2 Performance and Differences of Bias-ply and Radial Technologies [15]….. 12

Table 2.3 Comparison of Wireless network parameters that used in In-vehicle transmission [24]…………………………………………………………………….…... 18

Table 2.4 Energy Consumption of several wireless standard in different stages [26]……………………………………………………………………………….. 19

Table 2.5 The Measured RSSI according to sensor position [35]……………………. 25

Table 2.6 Variation of packet received at different speed [36]………………….……. 27

Table 2.7 Compilation of Limitation, Gap and Further Improvement from Research papers ……………………………………………………………………………………. 28

Table 3.1 Comparison of absolute pressure sensors ………………………………..... 37

Table 3.2 Comparison of IMU Sensors [38]…………………………………………... 38

Table 3.3 Comparison of wireless technologies [39]………………………………..… 39

Table 3.4 Comparison bluetooth module available in the market [40]…………..…. 40

Table 3.5 Comparison of Arduino Version Available In Market [41]…………...…. 41

Table 3.6 Software Used in project development ………………………………...….. 42

Table 4.1 Accelerometer result of different speed ………………………………..….. 55

Table 4.2 Z-axis behaviour during brake ………………………………………………. 58

Table 4.3 Description of tyre running condition ……………………………………… 59

Table 4.4 Experiment results of rotation counter with different speed …………...... 60

Table 4.5 Differential Voltage reading respect to pressure rereading …………….... 66

Table 4.6 Parameter uses to calculate the equation of straight line ………………… 67

Table 4.7 Pressure sensor readings by reverse engineering method ……………….. 69

Table 4.8 RSSI from multiples transmitter direction ……………………………….. 72

Table 4.9 Transmitter position in rear and front tyre………………………………... 73

Table 4.10 RSSI with different rotation speed ……………………………………….. 75

LIST OF FIGURES

Figure 2.1 Sensor Placement for Direct System [9]……………………..………..….. 8

Figure 2.2 Operation Principle for Indirect System [12]………………..……….….. 8

Figure 2.3 Tyre Monitoring Technology Categories ………………………………... 9

Figure 2.4 Bias-ply construction [15]…………………………………………………. 11

Figure 2.5 Radial construction [15]…………………………………………………… 12

Figure 2.6 Description of radial tyre components [16]………………….……………. 14

Figure 2.7 Tyre Failures Caused by Improper Inflation [18]………………….…….. 15

Figure 2.8 Wear Bar inside the Tyre Thread [22]…….……………………………… 16

Figure 2.9 Calculation of Wheel Speed [28]………………….……………………….. 20

Figure 2.10 Measured and estimated vehicle speed according to time [29]………… 20

Figure 2.11 Frequency of rotation compared to oscillation [31]…………………….. 21

Figure 2.12 Traction ratio agianst slip different inflation pressure [32]……..…….. 22

Figure 2.13 Straight-line braking with respect to inflation pressure and acceleration [34] ……………………………………………………………………….. 24

Figure 2.14 The Position of TPMS sensors and monitoring device [35]………….… 25

Figure 2.15 LabVIEW GUI with speed variation [36]……………………….……… 26

Figure 3.1 Methodology of project …………………………………………………… 31

Figure 3.2 System flow of sensing and processing module …………………………. 32

Figure 3.3 Transmitter module sequence flow chart ……………………………….. 35

Figure 3.4 Receiver and Monitoring module sequence flow chart ………………… 36

vii

Figure 3.5 Selected Honeywell pressure sensor [37] ……………..…………………. 37

Figure 3.6 Selected IMU [38] ……………………………………..………………...... 38

Figure 3.7 HC-05 Bluetooth Module [40]……………….………….…….………….. 40

Figure 3.8 Arduino Pro Mini 328 [41]……………………………………………..… 41

Figure 3.9 Side view (left) and top view (right) of transmitter module placement . 44

Figure 3.10 Transmitter module (zoom in) …………………………………………. 45

Figure 3.11 (a) Transmitter module schematic design ……………………...... 46

Figure 3.11 (b) Transmitter module circuit diagram ……………………...... 46

Figure 3.12 (b) Receiver module schematic design ………………………………… 47

Figure 3.12 (b) Receiver module circuit diagram ………………………………….. 48

Figure 3.13 Pitch, roll and yaw behaviour …………………………………………. 49

Figure 3.14 Single axis accelerometer rotation …………………………………….. 50

Figure 3.15 Tyre dimension index [42] ……………………...... 51

Figure 3.16 Multiples direction of transmitter …………………………………….. 52

Figure 3.17 Multiples direction of transmitter …………………………………….. 53

Figure 4.1 The graph at 25 rounds per minute, rpm ……………………………… 56

Figure 4.2 The graph at 50 rounds per minute, rpm ……………………………… 57

Figure 4.4 Behaviour of Z-axis at tyre acceleration, sharp brake and stopped condition ………………………………………………………………………………. 59

Figure 4.6 Acceleration versus Time graph based on test runs …………………… 61

Figure 4.7 Technique for wheel rotation identification ……………………………. 62

Figure 4.8 Algorithm for rotation calculation ……………………………………… 63

Figure 4.10 Output reading show in terminal emulator program ………………… 64

viii

Figure 4.11 Histogram of Absolute Pressure versus Pressure Gauge Reading …… 65

Figure 4.12 Gauge pressure reading at 156 kpa …………………………………….. 68

Figure 4.13 Graph of Pre-set Pressure Reading versus Sensor Reading ………….. 70

Figure 4.14 Real-time pressure measurement on LCD …………………………….. 71 Figure 4.15 Re-position for central receiver ………………………………………… 74

LIST OF ABBREVIATIONS

ABS - Antilock Braking System

ATIS - Automatic Tyre Inflation Systems

CAN - Controller Area Network

CTIS - Central Tyre Inflation Systems DC - Direct Current

DSSS - Direct Sequence Spread Spectrum

FHSS - Frequency-Hopping Spread Spectrum

GUI - Graphical User Interface

HIL - Hardware-In-Loop

HR-WPAN - High-Rate Wireless Personal Area Network

IEEE - Institute of Electrical and Electronics Engineers

IMU - Inertia Measurement Unit

LCD - Liquid Crystal Display

LED - Light-Emitting Diode

LR-WPAN - Low-Rate Wireless Personal Area Network

MIROS - Malaysian Institute of Road Safety

NMVCCS - National Motor Vehicle Crash Causation Survey OEM - Original Equipment Manufacturer

PWM - Pulse Width Modulation

RF - Radio Frequency

RSSI - Received Signal Strength Index

TPMS - Tyre Pressure Monitoring System UDMS - Universal Data Monitoring System

USM – Universiti Sains Malaysia

WLAN - Wireless Local Area Network

PEMODELAN TENTANG KELAKUAN PUTARAN TAYAR MELALUI SISTEM PENGAWASAN TEKANAN TAYAR

ABSTRAK

Peningkatan bilangan kenderaan bermotor yang pesat di negara-negara berorientasikan teknologi telah membawa kepada peningkatan yang drastik dalam kemalangan jalan raya disebabkan oleh beberapa faktor. Faktor-faktor ini boleh dikategorikan kepada tiga faktor utama iaitu keadaan persekitaran jalan raya, tingkah laku manusia, dan masalah kenderaan. Di antara ketiga-tiga faktor, masalah kenderaan merupakan satu-satunya parameter yang boleh dimanipulasi apabila dibandingkan. Berdasarkan statistik, kajian mendapati keadaan tayar dan roda motosikal adalah punca utama yang menyumbang kepada kemalangan maut jalan raya. Oleh itu, adalah perlu untuk membina sistem yang dapat memantau keadaan tayar motosikal di jalan raya. Walaupun terdapat beberapa sistem pemantauan yang sedia ada, tetapi setiap sistem mempunyai kelebihan dan kekurangan tersendiri dalam kekangan aplikasi. Sebagai contoh, parameter utama seperti bacaan tekanan pneumatik dari tayar motosikal tidak dikemaskini secara langsung, hal ini boleh menyebabkan keadaan menjadi lebih teruk apabila berlaku kebocoran pada tayar. Selain itu, keadaan putaran roda seperti kenaikan dan penurunan pecutan serta kecekapan cengkaman brek yang tidak diambil kira boleh menjurus kepada penghasilan haba, terutamanya di negara-negara yang berada di garisan Khatulistiwa yang mempunyai jalan raya yang panas sepanjang siang hari. Di samping itu, penempatan pemancar dan penerima bagi tujuan komunikasi tanpa wayar perlu diperbaiki untuk memastikan kualiti penghantaran maklumat dapat dilaksanakan dengan baik, tindakan ini bertujuan untuk mengelak transmisi maklumat yang salah atau tertangguh. Objektif kajian ini adalah untuk membangunkan satu sistem pemantauan yang menggabungkan kelebihan sistem pengukuran secara langsung dan tidak langsung dalam usaha untuk mengatasi masalah seperti yang dibincangkan. Sistem ini perlu memantau bacaan tekanan tayar yang dikemas kini secara langsung dan membuat kiraan jumlah jarak perjalanan menggunakan algorithma berdasarkan kajian keadaan putaran roda kenderaan tersebut. Selain itu, parameter tahap kuasa telah dikaji melalui penunjuk kekuatan penerimaan isyarat untuk tujuan pemantauan kualiti transmisi. Sistem ini mempunyai dua bahagian iaitu modul pemancar dan modul penerima. Modul pemancar dibina daripada kombinasi perkakasan seperti pengawal mikro modul bluetooth dan peranti pengesan yang terletak pada rim tayar untuk memperolehi status keadaan tayar. Manakala modul penerima berfungsi sebagai pengumpul dan penganalisa maklumat yang diterima dari modul pemancar dan memberi maklum balas apabila status keadaan tayar tidak normal. Hasil keputusan daripada beberapa eksperimen yang telah dijalankan menunjukkan bahawa penempatan pemancar dapat memastikan bacaan penunjuk kekuatan penerimaan isyarat yang konsisten iaitu pada -70 dBm dengan kelajuan putaran tayar yang berbeza dan kedudukan pemancar yang berbeza dari jarak yang sama. Hasil kajian juga menunjukkan bahawa prestasi putaran roda dapat dikenal pasti dan menghasilkan anggaran jarak yang dilalui kenderaan berdasarkan kiraan jumlah jarak perjalanan. Selain itu, tahap tekanan pneumatik tayar telah dirumus dan ketepatan hasilnya telah dipastikan dengan kaedah kejuruteraan balikan sebanyak ± 20 kPa daripada nilai toleransi projek. Secara keseluruhan, penyelidikan ini telah berjaya memperoleh bacaan tahap tekanan secara langsung daripada roda yang berputar, membuat kiraan jumlah jarak yang dilalui berdasarkan kitaran putaran roda dan menempatkan pemancar dan penerima berdasarkan parameter tahap kuasa untuk memastikan kualiti transmisi.

xii

THE MODELLING OF TYRE ROTATION BEHAVIOUR WITH TYRE PRESSURE MONITORING SYSTEM

ABSTRACT

The number of motorized vehicles is rapidly increasing in the technology driven countries, and led to the dramatic increase in road accident. The causes of accidents can be categorized into three major factors which are road environmental condition, human behaviour, and vehicle defects. The vehicle defects are the only parameter that is controllable when compared with to other two factors. Statistics show that the tyre and wheels-related from motorcycles is the critical reason and major contributor to road death accident. Therefore, there is the necessity to build a system that is able to monitor the on-road tyre condition. Several existing monitoring systems are available, but each has its own advantages and disadvantages based on the application’s limitation. For example, the important parameter such as pneumatic pressure captured from the tyre is not in real-time, thus it may become worst when there is air leakage. Besides that, tyre rotation behaviour such as acceleration, deceleration and sharp brake condition is not considered which may tend to build up heat. Especially in the countries on the equator which have warm road pavement throughout the daytime. In addition, the placement of transceiver for wireless communication need to determine in order to avoid misinterpretation on the wrong/delayed result captured. The research objective is to develop a monitoring system that combines the advantages of direct and indirect measurement system in order to overcome the problem as discussed. The system needs to capture the real-time pressure level on running tyre and provide calculations on the total distance travelled by the vehicle through algorithms from investigation of tyre rotation behaviour. Apart from that, the power level parameter was studied through the received signal strength index (RSSI) calibration for transmission quality purposes. The system consist of two parts which are the transmitter module and receiver module. The transmitter module is built from combination of hardware such as microcontroller, bluetooth module and sensing devices which sat on the tyre rim to acquire tyre condition. Whereas, the receiver module is responsible to collect and analyze information from the transmitter module and provide a feedback whenever an abnormal tyre condition occurred. Several experiments were conducted, the result shows that the placement of transceiver can be justified with consistent RSSI at -70 dBm from different tyre rotation speed and different transmitter’s directions with the same displacement. The result also shows that the performance of tyre rotation behaviour is able to identify and provide the estimation of distance travelled by the vehicle with evidence support from distance travel calculation. Lastly, the pneumatic pressure level inside the tyre was captured and the result accuracy is further ensured with reversed engineering method with ± 20 kpa from project tolerance. Overall, the research work is able to capture the real-time pressure level on running tyre, provide calculation on total distance travelled based on tyre rotation cycle and position the transceiver based on the power level parameter to ensure the transmission quality.

xiii

CHAPTER 1

INTRODUCTION

1.1 Background

In a technology driven century, the number of road accident is dramatically increase due to the rapid rising of motorized vehicles on the road as tabulated in the statistics of General Road Accident Data in Malaysia [1]. Table 1.1 clearly stated the trend of road crashes for 20 years until 2016. The number of road crashes in the year 2015 is nearly multiple with a factor of 1.5 which equal to 161,000 cases when compared to the year 2005. In addition, road deaths are weighted 6706 of cases from the road crashes, this remains highly unacceptable which results in very high economic and social costs to the nation.

Table 1.1 Statistic of General Road Accident Data in Malaysia [1]

According to the Malaysian Institute of Road Safety Research (MIROS) report, there are three major factors that contribute to road accidents, such as human factor, road environmental condition, and vehicle defects [2]. The vehicle defect contributed 6.2% from the factor, which is at huge amount of cases when converted into numberical value, 30,335 cases to be exact.

In order to root cause this issue, road deaths can be categorized by the type of vehicle involved as shown in Table 1.2. Motorcycles and motorcars are the major contributors in road death which weighted 83.5% among the road death crashes [2].

Table 1.2 Statistic of Road Deaths by Type of Vehicle [2]

Types of Vehicle The Number of Deaths Motorcycle 4485 Motorcar 1489 Pedestrian 511 Lorry 186 4 Wheel Drive 142 Others 122 Bicycle 123 Van 65 Bus 29 Total 7152

Besides that, vehicle defect is the only parameter that is able to be controlled and prevented before an accident happens when compared to the other two major factors. The vehicle defect can be further attribute into several critical reasons. From the publication of The National Motor Vehicle Crash Causation Survey (NMVCCS), the vehicle related critical reasons were mainly measured through external inspection of the vehicle components such as tyres, brakes, and steering as shown in Table 1.3. The table shows that the tyre/wheels-related components are the known reason which contributed the highest number of cases, weighted 35% from the vehicle defect factor [3].

Table 1.3 Vehicle Related Critical Reasons [3]

Based on the comparisons of statistical result, it is concluded that the major contributor to the number of accidents is the vehicle defects from motorcycle with tyre/wheels related issues. Therefore, there is a need to develop a reliable tyre pressure monitoring system that considers tyre rotation behaviour such as acceleration, deceleration and sharp brake condition, in order to reduce the number of accidents.

1.2 Problem Statements

The main purpose of the investigation of tyre rotation behaviour through modelling algorithm with tyre pressure monitoring system is to reduce the number of accidents in term of tyre related defects. In addition, the proposed algorithm for tyre rotation cycles can be utilized for developing driverless vehicle or smart car systems. There are several types of tyre monitoring system available and it can be categorized into two systems which is direct measurement and indirect measurement. In short, the direct measurement system measures the tyre pressure through sensing device, whereas indirect measurement system measures the other parameters other than tyre pressure. Both direct and indirect measurement has advantages and disadvantages based on working principle.

Existing monitoring systems are unable to show the real-time pneumatic pressure level inside the tyre when the vehicle is running and this may become worse whenever any air is leaking. The real- time system was proposed with the advanced integration techniques applied to provide real-time

3 tyre pressure monitoring [4]. Although the real-time reading can be obtained via integration techniques, it's only applicable to stationary vehicle wheels. None of the experiment is discussed during their rotation. Apart from that, the pneumatic pressure level inside the tyre will increase with respect to distance travelled due to several rotational behaviour such as acceleration and sharp braking which may lead to the buildup of heat, especially under hot ambient temperature.

Several great publications focused on enhancing the monitoring system either through hardware and software. There are hardware minimization through specific antenna design [5] and power recovery circuit with battery-less [6] while there are also software based implementations such as off-road simulation and graphic user interface [7]. However, none of these topics survey the placement and positioning of the transceiver. This is the important factor that may affect the data transmission quality and lead to misinterpretation on the captured result.

1.3 Research Objectives

The research objectives

i. To develop a monitoring system with the advantages of indirect measurement system, which provide calculation on total distance travelled by the vehicle. ii. To develop a monitoring system with the advantages of direct measurement system, which able to capture the real-time pressure level on running tyre. iii. To determine the placement of the transmitters and receiver based on power level parameter through the received signal strength index (RSSI) calibration to ensure the transmission quality.

1.4 Research Scope

From the statistical survey discussed in the previous section, vehicle defects of motorcycles related to tyre/wheels is the major contributor to the road death accidents, hence this research is aimed to provide the monitoring system based on the tyre condition. This monitoring system will focus on motorcycles tyre which is the highest number of vehicle from crash report cases [3]. The literature

4 review on the type of tyre will be discussed, only the tubeless radial tyre will be applied to the research work because of the tyre construction was suitable to be used in experiment while the other type of tyre is not considered. The tyre model for this research is fixed with 80/90-17 M/C 44P. Besides, gauge pressure is used in several algorithm calculations instead of absolute pressure because the experiment was not test in a vacuum environment. The research system monitored on the parameters such as pneumatic pressure inside the tyre, tyre temperature and wireless coverage throughout several experiments. From the experiment, pneumatic pressure will be monitored at range from180 kPa to 270 kPa with 20 kPa of tolerance and resolution, and temperature at 20 to 65 degree Celsius with reference to the weather range of Malaysia [8] and project coverage range with 0.1 degree Celsius of resolution. The Bluetooth coverage ranges is from 0 to 4 meters, which covers the motorcycle size. All the experiments will take place in standard road conditions with dry and flat surface (tar road) and altitude of 60 meters as the guidance in order to achieve the stated objectives with consistent results.

With literature surveyed and fixed project specification, this research will able to investigate the tyre rotation behaviour through modelling algorithm with the proposed system (direct and indirect measurement), which is small in size and able to function well inside the tyre. This research will use motorcycle (2-wheel vehicle) to develop the stated objectives such as the real-time pressure level, the total distance traveled algorithm and the placement for transceiver with multiple sensors (accelerometer, pressure sensor, and thermometer) mounted inside the tyre. The monitoring system is able to show the real time tyre condition in term of pneumatic pressure and temperature parameters and provide feedback system through display, light indicator and alarm when the tyre runs under abnormal condition.

1.5 Research Contribution

This research contributes to provide alternative distance traveled calculation based on the proposed algorithm through developing the tyre pressure monitoring system. The system was enhanced from only measure of pneumatic pressure to tyre rotation behaviour such as acceleration, deceleration and sharp braking condition. Apart from that, the placement for the transceiver is relocated with the determination of RSSI to ensure the transmission quality.

1.6 Thesis Outline

In chapter 2, the results of comprehensive literature review is deliberated with more in-depth counterpart of the literature review in chapter 1. The overview of various segments such as existing tyre pressure monitoring system, tyre construction and inflation is discussed. The paper review on several techniques comparison for tyre rotation behavior, wireless technologies applications, and the relationship between pneumatic pressure level and tyre performance is conducted.

In chapter 3, the methodology of the research is charted on how the research objective can be achieved. The five development stages of this research work will be discussed. The system flow of sensing module is presented in terms of a flow chart in the first stage, continued with comparison and selection of hardware, software and sensing device. The third stage is the proposed system circuit design and implementation on motorcycle rim. Next, the experimental setup for each experiment is discussed in detail with specific situations and theories involved.

In chapter 4, the discussions and findings from each experiment will be conducted in this chapter. The result obtained from the system is presented in the form of a table or graph and followed by discussion on findings supported by theory, and analysis of the findings is done to determine whether the findings align with the experiment hypothesis assumption or vice versa contradict to the expected result.

In chapter 5, the nutshell of research work which summarizes and explains the aims, important findings and conclusion. The evaluation of modelling algorithm with commentary on the contribution and limitations of the research work was discussed. Lastly, recommendations for further improvement areas that are needed to enhance the performance and accuracy of the system is discussed.

CHAPTER 2

LITERATURE REVIEW

2.1 Introduction

The overview of existing tyre pressure technologies was discussed in terms of the advantages and disadvantages based on application. The tyre construction and inflation conditions such as over inflation and under inflation is deliberated. This literature review is necessary before developing the system due to its specific application, for example, existing tyre technologies Central Tyre Inflation Systems (CTIS) are only suitable for bus/truck tyres due to the operation principle. Several proposed techniques from the paper had been reviewed to make comparisons and areas that needed improvements were identified. The review on tyre rotation behavior, wireless technology applications, and the relationship between pneumatic pressure level and tyre performance is conducted.

2.2 Existing Tyre Monitoring Technologies

2.2.1 Direct and Indirect Measurements

The tyre monitoring system is developed to acquire pneumatic pressure level reading inside the tyre for specific vehicle based on several technologies. These monitoring system was classified into two major categories such as direct and indirect measurement [10]. The direct and indirect measurement have different operation principle, but both measurements are able to provide feedback when the tyre runs under abnormal pressure condition. The direct systems will attach a sensing devices and a transmitter inside the tyre as illustrated in Figure 2.1 and transmit the information wirelessly to the receiver. The information was analyzed by the system and warns the driver if the tyre pressure is below or above predetermined level. The direct systems are able to detect pressure levels as small as one PSI (pounds per square inch) in term of resolution. Whereas,

7 the indirect systems have an alternative way in monitoring the tyre instead of checking the pneumatic pressure level. This system monitors the rate of revolution from each wheel. For example, as shown in Figure 2.2 the tyre that has lower pressure will roll at a different revolution per distance as compared to other tyres. The system will feedback the abnormal tyre condition to the driver without generating the accurate pressure reading. The limitation of this system occurred when all the tyres runs under abnormal conditions and will result in misinterpretation of information [11].

Figure 2.1 Sensor Placement for Direct System [9]

Figure 2.2 Operation Principle for Indirect System [12]

2.2.2 Survey and Comparison of Tyre Monitoring Technologies

According to the survey conducted, there are five different types of monitoring approaches available with current technology. The available approaches are Tyre Pressure Monitoring System (TPMS), Central Tyre Inflation Systems (CTIS), Automatic Tyre Inflation Systems (ATIS), Dual Tyre Pressure Equalizers and Passive Pressure Containment Approaches as shown in Figure 2.3. Each technology addresses specific vehicle inflation problem. The comparison and description of the working principle of each technology was discussed in Table 2.1.

Tyre Pressure Monitoring System (TPMS)

Dual Tyre Pressure Central Tyre Inflation Equalizers Systems (CTIS)

Tyre Monitoring Technology

Automatic Tyre Passive Pressure Inflation Systems Containment (ATIS)

Figure 2.3 Tyre Monitoring Technology Categories

Table 2.1 The Comparison of Tyre Monitoring Technologies Working Principle [13, 14]

Pros Cons Tyre Pressure Monitoring System (TPMS) - Working Principle: Direct measure pressure level and compared with pre-set value. i. Direct measurement system i. Sensing device attached to fragile valve ii. Feedback system provided ii. Pre-installation necessary and the system is not standardized Central Tyre Inflation Systems (CTIS) - Working Principle: User owns the control and able to select the target pressure level in order to adjust the pressure level for specific operation i. Reduce the vibration and shock i. Not able to show the actual pressure level loading ii. Only used for off-road transport vehicle ii. Possible of flexible control Automatic Tyre Inflation Systems (ATIS) - Working Principle: Monitor tyre inflation level with a pre-set value and inflate/relief whenever the tyre is underinflated/overinflated. i. Automatically re-inflate tyre to pre- i. Not able to show the actual pressure level set pressure level ii. Rely on compressed-air tanks as inflation ii. Able to relieve pressure when over- source which occupied space inflated Dual Tyre Pressure Equalizers - Working Principle: Attempt to bring the same pressure level inside the tyre when facing any unequal loading, temperature, and slow air seepage. i. The track is leaking with visual i. Only used for truck or vehicle with dual display tyres ii. Balancing for both tyre pressure ii. Sensor mounted on hose connection levels between each tyre valve stem Passive Pressure Containment Approaches - Working Principle: Another medium inserted into the tyre and capable of maintaining the pressure level once inflated. i. Able to reduce natural air loss with i. Can mitigate the effect of punctures lower permeation rate ii. Provide barriers to air loss

2.3 Type of Tyre Construction, Improper Inflations and Tyre Depth Measurement

2.3.1 Bias-ply and Radial Tyre Construction

This section discussed the performance and differences between bias-ply and radial tyre construction, while other types of tyre were excluded such as summer tyre, winter tyre, and wet weather tyre. The construction method of bias-ply was shown in Figure 2.4 while radial was shown in Figure 2.5. Bias-ply versus radial tyre was tabulated in Table 2.2 in term of differences, contact to ground, temperature and cornering.

Figure 2.4 Bias-ply construction [15]

Figure 2.5 Radial construction [15]

Radial tyre is better compared to bias-ply tyre as it eliminated the unnecessary characteristic from bias-ply tyre. The radial tyre having lesser layers of body cord on its sidewall allows better flexibility. The thread can have a full contact area with the ground when experiencing cornering or heavy load. The bias-ply design is more independent as the sidewall and thread works separately with better than bias-ply permits. Therefore, the research work will only focus on the radial construction tyre shown in Figure 2.6.

Table 2.2 Performance and Differences of Bias-ply and Radial Technologies [15]

Bias-ply Radial Construction Method Bias-ply tyres in constructed into a single unit Radial tyre constructed of 2 parts which is one by layers of rubber coated with plies of about layer of rubber-coated with steel cables and the 30 degree angle of the diagonal. arc bead to bead with 90 degree angle. Pros and Cons from the construction i. Tyre thread will distort when i. Less distortion of tyre thread as the tyre experiencing heavy load due to the sidewall is flexible even when heavy deflection of sidewall. These will load applied. Tyre resistance to reduce the tyre life as decreasing the puncture is increasing with the vertical traction. deflection. ii. The performance of cornering is ii. More stable and balance when the tyre weaker when compared to radial due to is cornering because the sidewall and the strength of tyre sidewall. thread able to maintain the tread flat. iii. Increasing the layer of plies and bead iii. Increasing the diameter of steel cable cable wire able enhanced the used which preventing the tyre from strengthen hence reduce in chances of puncture and provide a cooling puncture mechanism as steel cable distribute iv. The drawbacks when the plies layer heat faster. increase is the built up heat due to the iv. The drawback when applying larger increase of mass, therefore resulting in diameter cable resulting in higher reduce tyre life. petrol consumption due the heavy weight.

Figure 2.6 Description of radial tyre components [16]

2.3.2 Tyre Failure Caused by Improper Inflation

Based on the survey done in chapter 1, the major contributor of accidents is tyre/wheel related issues from vehicle defect. This section will cover the critical defect of tyre failure caused by improper inflation. Tyre inflation failures can be categorized into 3 parts which consist of over- inflation, under-inflation and tyre wear [17].

Tyre over-inflation can defined as when the tyre experiences excessive pneumatic pressure. Due to the ride harshness was increased, the overinflating of tyres will result in serious tyre damage caused by any potholes or small sharp objects on the road. The tyre comes into contact with the ground at only the center portion. The small contact area increases the rate of wear and tear at tread

14 center and becomes more susceptible to any impact damage as illustrated in Figure 2.7. On the contrary, tyre under-inflation means the pneumatic pressure level inside the tyre is much lower than original equipment manufacturer (OEM) recommended operating inflation conditions. Under-inflated tyres when run with serious high temperature may lead to sudden blowout, especially during high revolution when a tyre is under inflated. Under inflation normally is due to a lack of frequent maintenance and slow air leakage from the tyre. Under inflation will result in excessive flexing of the sidewall, rapid wear of the tread shoulders, and high fuel consumption due to excessive friction between the tyre and ground surface as illustrated in Figure 2.7.

Figure 2.7 Tyre Failures Caused by Improper Inflation [18]

Frequent use of the vehicle or speeding may result in tyre being worn out Tyre wear means the reducing of tyre thread until it is lower than the acceptable tread depth which is 1.6mm [19]. Every tyre has the average life of about 30,000 to 60,000 km based on the various sizes and kinds of vehicle [20]. Through the accumulated mileage traveled by the tyre, the wear bars inside the tread grooves are seen which indicates wear condition as shown in Figure 2.8. The tyre is considered worn out when wear bars are flushed with the tyre tread. The other method of tyre wear

15 measurement can done by inserting the pinhead of tread depth gauge, which is the common and most accurate among the other measurement such as Penny Test and 20p Test [21].

Figure 2.8 Wear Bar inside the Tyre Thread [22]

2.4 Wireless Technologies

2.4.1 Overview of Technologies Survey in Wireless Communication for Vehicle

Wireless communication is the transfer of information or power between two or more points that are not physically connected. For TPMS, the transmission is done by a single or multiple transmitter(s), Tx and receiver, Rx. The wireless technologies were classified into a standard by the Institute of Electrical and Electronics Engineers (IEEE) such as IEEE 802.15.1, IEEE 802.15.3, IEE 802.15.4, and WiFi.

IEEE 802.15.1 or known as Bluetooth uses the FHSS technique (Frequency-Hopping Spread Spectrum), which splits the frequency band of 2.402-2.480 GHz into 79 channels (called hops) with 1 MHz for each channel. The signal is transmitted using a sequence of channels known to both transmitter and receiver. Therefore, by switching channels Bluetooth standard can avoid interference with other radio signals [23].

IEEE 802.15.3 is designed to facilitate High-Rate Wireless Personal Area Network (HR-WPAN) for fixed, portable and moving devices, IEEE 802.15.4 addresses the needs of Low-Rate Wireless Personal Area Networks (LR-WPAN) which is designed to facilitate those wireless networks, which are mostly static, large, and consuming small bandwidth and power [24]. Both of these standards use Direct Sequence Spread Spectrum (DSSS), and it does not allow changes of operating channels once a connection is initiated [25].

Wireless Local Area Networks (WLANs) also known as WiFi is based on the IEEE 802.11 standards and depending on local authority restrictions IEEE 802.11 b/g/n supports up to 14 channels in the 2.4 GHz frequency range. WiFi is a well-established network, wide spread and used in various environments and devices. The wireless network operates with three essential elements that are radio signals, antenna and router. The radio waves are keys which make the Wi- Fi networking possible.

2.4.2 Operation Mode

Wireless network can be categorized into two modes of operations which is Ad Hoc and Infrastructured [26]. The network which does not rely on a preexisting infrastructure, for example the routers in wired networks or the managed access points are known as ad hoc. Whereas the Infrastructured operation mode requires a base station that act as a central node to connect the wireless terminals. The base station provides the features to enable access to other wireless networks or the internet or intranet and wireless terminals use the base station to relay their messages. There is a drawback of this mode of wireless network, the wireless terminals will fail to communicate when the center point malfunctions.

2.4.3 Frequency, Data Rate and Range

Radio frequency (RF) is the electrical oscillations in term of electromagnetic wave frequencies that lie in the range extending from around 3 kHz to 300 GHz [27], which include the frequencies used for communications or radar signals. The data transfer rate is affected by the selected frequency, whereas the power consumption is rely on range covered. The comparison of the Wireless network parameter is tabulated in Table 2.3.

Table 2.3 Comparison of Wireless network parameters that used in In-vehicle transmission [24]

Standard Bluetooth High rate Low rate WiFi WPAN WPAN IEEE Spec. IEEE 802.15.1 IEEE 802.15.3 IEEE 802.15.4 IEEE 802.11

Frequency band 2.4 GHz 2.4 GHz 868/915 MHz ; 2.4 GHz ; 5 GHz 2.4 GHz Max. Data Rate 1 or 3 Mbps 11 – 55 Mbps Depend on 54 Mbps application App. Range < 10 m < 10 m < 20 m < 100 m Power Level 1 mA - 60 mA <80mA 20 μA - 50 μA ~ 116 mA Issues

2.4.4 Power Consumption

Power consumption of wireless technology is differentiated into 3 stages, such as transmit, receive and idle. The longer duration of wireless network is in idle the more efficient the power consumption. For TPMS where a battery source is the only the energy source to perform transmission, the power consumption of wireless technology must be taken as consideration. The energy consumption for several wireless protocols was tabulated in Table 2.4.

Table 2.4 Energy Consumption of several wireless standard in different stages [26]

Protocol Energy Consumption Sleep Transmit Receive ZigBee 0.06 μW 36.9 mW 34.8 mW Bluetooth 330 μW 215 mW 215 mW WiFi 6600 μW 835 mW 1550 mW

2.5 Review on Tyre Rotation Behaviour Model

Tyre rotational behavior under different speeds in terms of revolution and different inflation condition will be conducted clearly in this section. The data transmission between transmitter and receiver will be investigated. All parameters such as revolution, inflation and data transmission needs to be considered because the result obtained from the experiment is fully controlled by these parameters. For example, the tyre running in acceleration will affect the reading of y axis from the inertial measurement unit.

2.5.1 Tyre Rotation under Different Revolution

Tyre rotational behaviour can be categorized into 3 parts which are acceleration, deceleration and sharp braking condition. Dadashnialehi and his team published paper of Antilock Braking System (ABS) for In-Wheel Electric Vehicles using data fusion in 2013 [28]. The researchers improved the wheel speed measurement by the fusion concept with proposed novel architecture. ABS sensor is used in the measurement of wheel rotation speed by modulating the speed signal due to the frequency of the sensor which is influenced by the rotation speed. The concept of speed calculation is shown in Figure 2.9 with the reference clock applied, based on the angular velocity relationship to a radius of wheel and number of the gear teeth.

Figure 2.9 Calculation of Wheel Speed [28]

In the year 2015, Tannoury and his team introduced the variable structure observer for estimation of tyre rolling resistance and effective radius [29]. The paper proposed to consider the physical model of longitudinal dynamics and rotational speed of the wheel for wheel angular velocity and vehicle speed measurement. The test was carried out by the latter signal acquired from modern vehicle controller area networks (CAN). By the help of Newton’s second law, the rotational speed can be traced for the forces acting on the wheel and the results show that the measured reading is aligned with the estimated speed as in Figure 2.10. This approach have the additional works that convert the force applied to the vehicle speed by applying Newton’s law [30].

Figure 2.10 Measured and estimated vehicle speed according to time [29]

Bhuiyen and his partner introduced the low cost digital stroboscope for speed measurement. The measurement method for rotational speed of wheel is fully described [31]. The operating principle of the proposed strobe circuit will compare the reference frequency of oscillation and targeted rotational frequency. The reference and targeted frequency will have difference at first and the manually tune (from lowest to highest speed) the speed of reference oscillation frequency to synchronize with the targeted rotating substance frequency as shown in Figure 2.11. When the rotation speeds are parallel, the circuit will then capture and analyze the rotational speed by RPM formula. This proposed research work necessary to have the target rotating speed and take longer time on synchronization and speed measurement.

Figure 2.11 Frequency of rotation compared to oscillation [31]

2.5.2 Tyre Rotation under Different Inflation

Hendy and his team introduced the tyre pressure control system with LABVIEW program that is able to adjust and balance the pneumatic pressure level inside the tyres when different load is applied [32]. The adjustment of pressure range is between 1.15 bars to 2.25 bars (115 kPa to 225 kPa). The control algorithm is done by the 6-Rotary valve, the valve is used to inflate and deflate the air inside the tyre during rotation with the specific connection to the pressure line or atmosphere respectively. The net traction ratio (the act of pulling) against slip results in Figure 2.12 show that the tyre rotation behavior is directly influenced by the tyre inflation level. The tyre at high inflation level experienced less friction due to less tyre surface area contact to the ground and hence require low net traction. In addition, when heavier load is applied to the same inflation level tyre, the net traction significantly increases with respect to the same slip. The comparison of the tyre rotation behavior with a different inflation level is show in Figure 2.12 [32].

Figure 2.12 Traction ratio agianst slip different inflation pressure [32]

Another work was published in 2016 with self-inflating system. The inflation and deflation mechanism applied was similar with ‘Tyre pressure control system’ as discussed using the solenoid valve. This research has better covered inflation range from atmospheric 0 kPa until 500 kPa [33]. The tyre inflation level is measured by offset reading from the pressure sensor after conversion process as:

푉표푢푡 = 푉표푓푓 (푚푉) + 푆푒푛푠𝑖푡𝑖푣𝑖푡푦 (푚푉 / 퐾푃푎) ∗ 푃 (퐾푃푎) (2.0)

Which includes the sensitivity of the sensor. The experiment result obtained show that the Vout is directly proportional to pressure level as expected.

Shyrokau and his partners analyzed the subsystems coordination during straight-line braking with the tyre pressure inflation system using Hardware-in-loop (HIL) test rig [34]. The proposed test rig consists of hardware and software portions. The hardware consists of the brake system with hydraulic operation and tyre inflation pressure system, whereas the software includes MATLAB/Simulink for tyre motor simulation and multi-body vehicle model from commercial IPG CarMaker. The physical case study of straight-line braking is done with considering the initial velocity at 90 km / h on pavement with low friction coefficient. The result of the case study is shown in Figure 2.13 with respective parameters. From the plotted graph, it clearly shows that the tyre inflation pressure and longitudinal acceleration is directly influenced by the tyre rotation behaviour (straight-line brake). The vehicle acceleration is opposed to the action of brake as expected as braking action will slow down the revolution of a rotating wheel, whereas the inflation pressure level show contradictions which decreased from approximately from 3.5 bars to 2 bars when the action of the brake, this is due to the experimental setup. The tyre inflation level will increase with the action of braking if the tyre is stationary or runs in slow speed.

Figure 2.13 Straight-line braking with respect to inflation pressure and acceleration [34]

2.5.3 Data Transmission in Vehicle

In year 2016, Chul Kyun and his team proposed the tyre position registration algorithm with data of acceleration and received signal strength indicator (RSSI) [35]. The researcher determined the signal transmit position by using the acceleration data and the RSSI information obtained from receiver module. The result shows the received signal strength is directly proportional to the distance. The monitoring device is placed 1 meter apart to front tyre and 2.5 meter apart from rear tyre at the central of vehicle as illustrated in Figure 2.14. The experimental results show that the front tyres RSSI value is higher than rear tyre’s RSSI due to power of receiving signal is reduced gradually as the transmission distance is increased. This is to ease the estimation of sensors position using the RSSI. The acquired RSSI is stored in register through 4 bit Analogue to Digital converter, and the corresponding register is shown in Table 2.5.

Table 2.5 The Measured RSSI according to sensor position [35]

Figure 2.14 The Position of TPMS sensors and monitoring device [35]

In year 2015, Sharma and his partner discussed the effect of wheel spin velocity on the propagation of signal in TPMS was published [36]. By using a LabVIEW based implementation of automotive system as shown in Figure 2.15, the effect of the different speed of transmission and propagation was observed. The simulation of packet received against the different speed and distance between Tx and Rx was conducted. From the tabulated result as shown in Table 2.6 it is shown that the number of packets received is indirectly proportional to the revolution speed and transmission distance. Higher revolution speed and longer distance will result in packet loss during data transmission.

Figure 2.15 LabVIEW GUI with speed variation [36]

Table 2.6 Variation of packet received at different speed [36]

2.6 Chapter Summary

This section summarizes all the literature reviews related to the investigation of tyre rotation behavior through modeling algorithm of Tyre Pressure Monitoring System. The existing tyre monitoring technologies were surveyed and compared under direct and indirect measurement categories which consist of Tyre Pressure Monitoring System (TPMS), Central Tyre Inflation Systems (CTIS), Automatic Tyre Inflation Systems (ATIS), Dual Tyre Pressure Equalizers and Passive Pressure Containment Approaches. Based on the comparison, the TPMS is the best as it provides the critical feature in direct measurement and feedback system. The only drawback for the system was that the sensing device is attached to a fragile valve. The drawback was countered with the new proposed system that do not necessarily has to be attached to the valve with measurement guarantee.

The survey of tyre construction shows that radial tyre was chosen based on the comparison on pros and cons versus bias-ply tyre. Radial tyre is more suitable to be used in the experiment that tests on tyre rotation behavior based on the characteristic such as sidewall flexibility, and full contact area to the ground surface. The tyre inflation level was studied in terms of tyre under inflation, over inflation and tyre worn which will directly influence the tyre rotational behavior. Wireless transmission is the only path to convey information because wired transmission is not suitable to

27 be used in the rotating wheel. Therefore the wireless communication that is used in vehicle application, such as Bluetooth technology, WiFi, and High/Low rate WPAN was studied based on the operation mode, frequency, data rate, range covered and power consumption. There are advantages and disadvantages of each wireless technology, Bluetooth is chosen because the parameters offered fulfilled tthe research requirement especially on the performance of data transfer rate.

The paper review of validating on the tyre rotation behaviour model was compared and discussed in the last section of Chapter 2. This section includes paper reviews of tyre rotation under different revolution, different inflation, and data transmission in vehicle. The limitation and gaps from the papers reviewed that required further improvement was tabulated in Table 2.7.

Table 2.7 Compilation of Limitation, Gap and Further Improvement from Research Papers

Proposed Techniques Limitation & Gap Improvement area Data fusion [28] Reference clock is necessary in Replaced the encoder reference order to compare with the counter with alternative technique captured directly obtained input from sensors Variable structure The vehicle speed measurement Consider the measurement from observer [29] only tested at 40 km / h, experiment slow to fast speed with different speed was excluded

Digital stroboscope Synchronization for the targeted Improvement the possibility of 3- [31] object necessary before speed dimension speed measurement as measurement and only works for 2- tyre will in diagonal position dimensional when cornering Control system with The result is only obtained respect to Considered the critical parameter LabVIEW [32] difference load applied and with such as tyre rotation behaviour limited inflation range which directly influenced the inflation level under widen the inflation range

Self-inflating system The sampled data have a big interval Used better predefined pressure [33] of 55 kPa which out of the suggested sensor to improve the ± 20 % inflation based on 200 kPa performance in term of accuracy (motorcar)

Tyre position The experiment was conducted Conduct physical experiment registration algorithm under stationary condition with with rotating wheel to exclude the [35] proof of theory, result may be varies assumption made if by rotating wheel

Wheel spin velocity Fully depend on sensor and sensor Investigate on the placement for on propagation of location is not discussed in detail sensor and receiver with high signal [36] received signal strength index

CHAPTER 3

METHODOLOGY

3.1 Introduction

The implementation of tyre pressure monitoring system is discussed in the chapter. The new system is built with combination of both transmitter and receiver modules. The selection of hardware and software involved in the implementation were deliberated. The overall flow of the system is discussed in detail with the project methodology and flow chart. The system is designed to carry out the experiments that targeted the research objectives. The experiment setup for modelling of vehicle rotation behaviours, pressure sensor reading verification via reverse engineering method and the study of received signal strength index for ideal system placement is descripted in details. Explanation of the analysis procedure is discussed to justify on the sensor data captured throughout the experiments.

3.1 Methodology of Project

Figure 3.1 shows the working principle of the tyre pressure monitoring system. In detail, several tyre rotation behavior such as acceleration, deceleration, sharp braking, and tyre inner pressure level will be considered. These behaviors act as the project inputs which are acquired by the pre- installed sensing module inside the motorcycle’s tubeless tyre. The acquired raw data will be analyzed and feedback to the system will be provided when the tyre runs into abnormal condition. Bluetooth (IEEE 802.15) will act as the transmission line between the transmitter (Tx) and receiver (Rx) because the sensor was implemented inside the tyre, therefore wireless communication is necessary. The received signal strength index was monitored in order to ensure the quality of data transmission.

Sensing Pre-Processing Tyre Rotation Transmitter Behaviour Module Module

Analysing Feedback Processing Computed Receiver System to User Module Data

Figure 3.1 Methodology of project

The research work monitors the tyre rotational behaviour as well as the inner pressure level periodically through the sensing and processing module. The experiments had to:

 Investigate and model the tyre rotational behaviour such as vehicle acceleration, deceleration, and sharp braking.  Estimate the wheel counter based on the tyre rotation behaviour modelling.  Determine the positioning for transmitter and receiver to perform the better quality of transmission with received signal strength index.  Provide a direct sensing and monitoring system on tyre condition.

3.2 Block Diagram of Tyre Pressure Monitoring System

Figure 3.2 describes the overall flow of transmitter and receiver module that are linked together. At first, all the sensors, microcontroller A and Bluetooth master module (transmitter) were installed inside the tubeless tyre and seat on the wheel rim as shown in Chapter 3.5 The microcontroller A and B featured as the brain of the system which control all the processing activities inside the system respectively. Microcontroller A will be responsible for the activities before data transmission while microcontroller B is responsible for the activities after the data is received. Bluetooth modules will act as the transmission channel between both transmitter and

31 receiver parts. Two bluetooth modules are used in the research work, the transmitter (Tx) is the master module which is located inside the tyre, whereas the receiver (Rx) acts as slave module, located on the driver dashboard. The operation starts with capturing the input parameters from several sensing devices such as pneumatic pressure reading inside the tyre, wheel rotating speed, and tyre temperature. The acquired reading from the sensor was modulated by microcontroller A before sending. The transmission occurs once both modules get paired. The master module start transmitting the modulated data while slave module will demodulate the data before further processing. Microcontroller B combines and analyzes the sensor’s data and provide the actual tyre condition through the physical interface which is the displayed. The display, buzzer and LED light works together as a feedback system to the user when the tyre is running under abnormal condition.

Figure 3.2 System flow of sensing and processing module

3.2.1 Transmitter Module

The transmitter module consist of microcontroller A, bluetooth master module, batteries and a numbers of sensing device such as pressure sensor, accelerometer, and thermometer. The transmitter module is the module used to acquire the different parameter sensor readings from each sensor and send the processed data by microcontroller A via a master module. All the sensing devices are powered by the battery direct current (DC) source. All the sensors acquire the tyre condition continuously according to the configured sampling time. The sampled data will go through microcontroller A for processing before it is transmitted from the bluetooth master module. Bluetooth master and slave will pair up automatically once powered on, these are done by configuring the bluetooth device address since every bluetooth device will have a unique address and passcode for security purpose that avoid other devices trespass.

3.2.2 Receiver Module

The receiver module is a feedback system to user, it consists of liquid crystal display (LCD), light emitting diode (LED), buzzer, acknowledgement button and batteries. The module is able to show the tyre pressure and temperature level according to the sampled rate through the LCD. Furthermore, it also triggers the user when the tyre is under abnormal or dangerous condition. The buzzer and LED acts as an alarm and warning sign respectively in the feedback system. Users can also mute the buzzer after noticing the tyre situation by triggering the provided acknowledgement button. The display will continuously show the tyre condition by default sampling period.

3.3 Flow Chart

The overall sequence flow chart for transmitter module is shown in Figure 3.3. The system will begin when power supply is available, in this case the batteries source. All the installed sensor will start acquiring the parameter according to the tyre condition. The captured sensor’s readings are analyzed and modulated by microcontroller A before the transmission process. At the same time

33 the bluetooth master module will check the availability of the bluetooth slave module. If the bluetooth slave module is powered on, both bluetooth modules will automatically pair (as discussed in previous section 3.2.1) else bluetooth master module will continue to check the presence of bluetooth slave module. The transmission will occur once both modules are paired and Tx will start sending packets while Rx receives information.

The overall sequence flow chart of receiver module is shown in Figure 3.4. The function of the module is to analyze the tyre condition and provide feedback to the user. The received information from the bluetooth slave module will act as the inputs for Microcontroller B to determine the condition of the tyre. In this system, there are two conditions available which is Good and Danger condition. Good condition defined as pressure and temperature level inside the tyre within the acceptable range, while Danger condition is vice versa for example tyre run under/over-inflation. After the analyzing process through Microcontroller B and if the sensor reading is within the acceptable range, the system will print ‘Good Condition’. Contrary, when the sensor reading is out of the acceptable range, the system will print ‘Danger Condition’ and trigger the LED and buzzer as warning signs. At this state, the user can manually trigger the provided acknowledgement button to switch off the buzzer while LED lights ON remains or both buzzer and LED light will automatically switch off once the tyre condition return into an acceptable range. The monitoring system is switched on until the power source is cut off.

Figure 3.3 Transmitter module sequence flow chart

Figure 3.4 Receiver and Monitoring module sequence flow chart

3.4 Selection of Hardware

3.4.1 Pressure Sensor

Table 3.1 shows the comparison of absolute pressure sensor in market survey. The stand out sensor was manufactured by Honeywell as shown in Figure 3.5 because it offers a better operating pressure range which is suitable for vehicle pneumatic pressure measurement. Apart from that, the other specifications fulfilled the research scope, such as operating voltage, sensitivity and operating temperature. Although the sensitivity is not among the best, but is does not directly influence the system because system is having ± 2 psi (13.7895 kPa). Therefore the sensitivity is within the acceptable range. In addition, the sensor is cheapest in price when compared to other manufacturers.

Table 3.1 Comparison of absolute pressure sensors

Product Price, Pressure Voltage, Sensitivity, V / Temperature, Brand RM range, kPa volts Pa Celsius Honeywell 23.00 0 - 413 5.0 21 mV / kPa -40 to 125 Freescale 44.50 0 - 200 10 - 16 0.2 mV / kPa -40 to 125 VTI 200.00 30 - 120 3.3 Resolution: 1.5 On-chip Pa All sensors 420.00 60 - 413 4.5 - 5.5 Not stated -40 to 125 Infineon 28.00 60 - 165 5 43.8 mV / kPa -40 to 125

Figure 3.5 Selected Honeywell pressure sensor [37]

3.4.2 Inertia Measurement Unit (IMU)

There are various kinds of combos IMU available from the market survey. Table 3.2 shows the models that consist of the combination of sensors such as accelerometer, magnetometer, and gyro. The selected IMU is shown in Figure 3.6. Both IMUs consist of four sensing devices that fulfilled project requirement with provided I2C. The reason GY-80 BMP085 9-Axis Magnetic Acceleration Gyroscope Module was selected because it cheaper in price compare to another IMU and compatible with Arduino.

Table 3.2 Comparison of IMU Sensors [38] Triple Axis Accelerometer, ±2000 ° / s 9-Axis Magnetic Acceleration Gyro, Barometric and Magnetometer Gyroscope Module for Arduino USD $39.50 USD $13.21 Consist of: Accelerometer; Gyro; Consist of: Accelerometer; Gyro; Magnetometer Magnetometer Compatible to Arduino boards Compatible to 3 and 5 volts system

Figure 3.6 Selected IMU [38]

3.4.3 Wireless technologies Table 3.3 shows the list of wireless technology that consists of Bluetooth, Zigbee and WiFi. Each IEEE protocol has its own distinctive advantages and limitations. The variation in the application may expand the potential developments. Bluetooth classic stand out among technologies for this research due to higher data rate compared to Zigbee. The data rate is important becuase there is huge amount of information captured from multiple sensors, therefore high data rate needs to be considered. Bluetooth is suitable to be used in cable replacement applications. Bluetooth is also easier to configure when compared to ZigBee. Although WiFi provides excellent parameters in term of data rate, data throughput and maximum range, it is not suitable in the research work because the proposed system requires that not every location provides WiFi services. Apart from that, WiFi has a high power consumption application and is therefore not suitable for external power source application.

Table 3.3 Comparison of wireless technologies [39]

Name Bluetooth, Classic ZigBee WiFi IEEE Standard 802.15.1 802.15.4 802.11.x Frequency (GHz) 2.4 0.868, 0.915, 2.4 2.4 and 5 Maximum raw bit rate Depend on 1-3 0.25 (Mbps) technology Typical data throughput Depend on 0.7-2.1 0.2 (Mbps) technology 10 (class 2), Maximum Range (meter) 10 - 100 100 - 250 100 (class 1) Power Consumption Medium Very low High Example Battery Life Day Month Hour

3.4.4 Bluetooth

Based on project requirements, Table 3.4 compares two types of bluetooth modules specification survey. Wireless Bluetooth RF Transceiver Module TTL HC-05 stand out because of ease on configuration and setup. Apart from that, when it comes to comparing the price, the model is cheaper. The selected Bluetooth module is shown in Figure 3.7.

Figure 3.7 HC-05 Bluetooth Module [40]

Table 3.4 Comparison bluetooth module available in the market [40]

DF Robot Bluetooth Module Wireless Bluetooth Module HC-05 USD 25.00 USD 7.99 • Emission power: < 4 dBm • Emission power: < 4 dBm • Sensitivity: < - 84 dBm • Sensitivity: < - 84 dBm • Transfer rate: • Transfer rate: Asynchronous: 2.1 Mbps (Max) / 160 kbps; Asynchronous: 2.1 Mbps (Max) / 160 kbps; Synchronous: 1 Mbps / 1 Mbps Synchronous: 1 Mbps / 1Mbps • Connection encrypt • Connection encrypt • Profiles: Serial port • Profiles: Serial port

• Baud rate: 4800 - 1382400 • Power supply: + 3.3 VDC at 50 mA

• Input Voltage: 3.3 VDC at 50 mA • Operating temperature: - 20 to 75 celcius

3.4.5 Microcontroller The proposed system was implemented inside the motorcycle tyre, hence the sizing of the hardware needs to be considered. Arduino Pro Mini is the only model that is small in size when compared. The comparison of Arduino Pro Mini was listed in Table 3.5. Arduino Pro Mini 328 – 5 V / 16 MHz was selected as shown in Figure 3.8 because the higher speed will smoothen the complexity flow in receiver module which is used for monitoring and analyzing the multiple inputs.

Figure 3.8 Arduino Pro Mini 328 [41]

Table 3.5 Comparison of Arduino Version Available In Market [41]

Arduino Pro Mini 328 – 5 V / 16 MHz Arduino Pro Mini 328 - 3.3 V / 8 MHz $ 9.95 $ 9.95  16 MHz with external resonator  8 MHz with external resonator (0.5 % tolerance) (0.5 % tolerance)  Supports auto reset  Supports auto reset  5 V regulator  3.3 V regulator  DC input 5 V / 12 V  DC input 3.3 V / 12 V  Analogue/Digital Pins: 8 ; 14  Analog/Digital Pins: 8 ; 14  33 x 18 mm  33 x 18 mm  Weighs < 2 grams  Weighs < 2 grams

Table 3.6 show the list of software which is involved in the implementation of system. The list of software includes the hardware design, system program and signal verification. The particular use of each software is discussed respectively in detail as shown in Table 3.6.

Table 3.6 Software involved Software Detail Description Used Autodesk Inventor is a tool used for 3D mechanical or hardware design, Autodesk- simulation and documentation. Inventor This software is used to design Tyre Pressure Monitoring System (TPMS) in

detail and placement of system on rim with a better view and ease of

explanation. Arduino IDE is a software to write code and upload to the Arduino board. The Arduino-IDE environment is written in Java and other open-source software.

This software used to program and burn into an Arduino Pro Mini board that

act as the controller for the system as well as algorithm calculation. The terminal emulator program. It able to support several types of terminals Tera term and serial port connections. Teraterm is used for acquiring sensor data when experiment, due to the data unable to obtain via wire connection. In addition, it also used for bluetooth configuration. The open-source software services for Processing and Arduino that allows documentation, prototyping and layout design. Fritzing This software used to design the circuitry for both transmitter and receiver module.

The software for the bluetooth client communication tools (ie: bluetooth slave Bluetooth mode), bluetooth serial communication can be tested. SPP Tools This android application able to provide quick check the bluetooth connection Pro and display the acquired data. Z-DeviceTest check on Android device sensors in an intuitive and comprehensive way offering in-depth analysis all the characteristics of your Z- Device smartphone. Test This application helps to study the pattern of waveform with specified sensing devices.

3.5 Design & Experiment Setup for Tyre Pressure Monitoring System

In this research, the vehicle type and tyre construction type were specified as motorcycle and radial construction tyre.

3.5.1 Design of Transmitter Module inside the Tyre

Traditional tyre pressure monitoring system only consists of a single sensing device which is the pressure sensor, which mean the pressure level is the only input parameter and hence the system had limited application. The proposed system not only enhanced the features, but determine the better placement for the system. Enhancing the features of the system offered more research opportunity on tyre rotation behaviour. Furthermore, the better placement of the system plays important role in communications, which is the data transmission to the outside environment from the tyre.

The system requires miniature in size in order to fit inside the motorcycle tyre. Figure 3.9 (left) shows the implementation of the system and the way system seated or placed on the wheel rim. Compared to the traditional system that sensor must be mounted to the tubeless rim valve hole as shown in Figure 3.9 (right). The new implemented system is not necessary to be mounted on the valve hole but can be mounted at any point on the rim base along the circumference, hence the system’s flexibility was increased.

Proposed system

Valve hole

Motorcycle rim

Figure 3.9 Side view (left) and top view (right) of transmitter module placement

Figure 3.10 shows the overview of the transmitter module. The dimension transmitter module of 5 x 5 x 1.5 cm for length, width and height. From the zoom-in view clearly stated that the transmitter module was built with combinational hardware that consist of the Arduino Pro Mini, Bluetooth master module, power source and sensing devices such as pressure sensor and inertia measure unit (IMU). The IMU consists of accelerometer, magnetometer and thermometer, which is used for investigating the tyre rotation behaviour.

BT Master Module Power source

Sensing devices Rim Base Arduino Pro Mini

Rim Flange Arduino Pro Mini ini

Figure 3.10 Transmitter module (zoom in)

3.5.2 Transmitter and Receiver Module Circuit Design

Figure 3.11 (a) shows the schematic design of transmitter module and Figure 3.11 (b) shows the hardware and wiring connection that built the transmitter module, red colour and black colour wiring represent Vcc and Gnd respectively throughout the whole circuit. The majority of the hardware is powered by a DC battery source except for the bluetooth module which is connected to Arduino board 3.3 volt Vcc pin since it has lower operating voltage. The Arduino board acts as the central control of the whole module, its analogue pins is connected to the pressure sensor, while the digital pin is connected to the inertial measurement unit (IMU) due to different operating principle from each hardware. Bluetooth’s Tx and Rx pins are connected to Arduino’s digital pulse width modulation (PWM) pin and all hardware share the common ground.

Figure 3.11 (a) Transmitter module schematic design

Figure 3.11 (b) Transmitter module circuit diagram

Figure 3.12 (a) shows the schematic design of receiver module and Figure 3.12 (b) shows the monitoring system that is built from combination of hardware. Similar to the transmitter module.

The red colour wiring represents Vcc while the black colour represents Gnd and all the hardware share common ground. The module is powered by DC source rather than batteries, because the module is not placed inside the tyre as compared to the transmitter module which is inside the tyre. The potentiometer acts as a variable resistor used to tune the LCD to get a sharp visual. The LCD will show the tyre temperature and pneumatic pressure condition inside the tyre while the buzzer and LEDs act as a warning sign when a tyre is under abnormal conditions. The buzzer will be muted once the acknowledgement button was triggered which means that the user has acknowledged the tyre issue. LEDs remain lighted when tyre run in abnormal condition.

Figure 3.12 (a) Receiver module schematic design

Figure 3.12 (b) Receiver module circuit diagram

3.5.3 Experimental Setup

This section will discuss the experimental setup and analysis procedures in order to achieve the objectives as discussed in Chapter 1. Three experiments were conducted based on the modelling of vehicle rotation behaviours (acceleration, deceleration and sharp braking); pressure sensor readings verification via reverse engineering algorithm; and the study of received signal strength index for ideal system positioning. The sampling rate for data acquisition was 100 ms throughout the overall experiments. All the experiments were conducted on a flat road surface/pavement.

The experiment on modelling of vehicle rotational behaviour is conducted with several apparatus such as transmitter module, motorcycle, terminal emulator program, and tachometer. The performance of accelerometer based on wheel rotation with speed variation, wheel at braking/stop condition is discussed. The experimental setup for these 3 sub experiments are identical. The transmitter module was installed inside the motorcycle tyre and seated on the surface the of tyre

48 rim. The Bluetooth master module at transmitter module and the Bluetooth slave module at receiver module was configured, so that it is paired automatically once powered. The tyre rotation condition was captured by the sensing devices with 100 sample data for every experiment. The data was recorded in the log file with the help of terminal emulator program. The experiment on accelerometer performance based on different speed variation at 25, 50, 75, and 100 rpm was carried out to determine the roll, pitch and yaw behaviour as shown in Figure 3.13. The roll, pitch and yaw behaviour of accelerometer which represent x axis, y axis, and z axis in the research.

Figure 3.13 Pitch, roll and yaw behavior

Besides that, the experiment on accelerometer performance at braking/stop condition was carried out. This experiment only determined the z-axis as it is representing the acceleration of wheel. Therefore z-axis was used for the determination of wheel rotation in terms of acceleration, deceleration and sharp braking condition. Five brake attempt tests was carried out to study the pattern of waveform from z-axis. In sharp braking condition, the quick deceleration over time taken shows the sharp slope in the z-axis. The result was aligned with the working principle of accelerometer as shown in the Figure 3.14. Figure 3.14 shows the method to obtain the tilt from the accelerometer. The tilted angle can be calculated by using the equation of,

퐴 α = arcsin (3.0) 𝑔

Where A is the acceleration and g is the earth gravity vector. Therefore, the sharp slope waveform is able to be generated during sharp braking condition.

Figure 3.14 Single axis accelerometer rotation

Furthermore, the estimation on distance travelled by the vehicle was tested. The distance travelled is the product of number of cycle rotations and circumference of wheel as stated in equation 3.1. The experiment is carried out with 50 cycles of rotation at 4 different speed applied. The number of cycle rotations is able to identify through the y axis of the accelerometer with array operator applied, whereas the circumference of wheel was fixed with tyre size throughout the experiment. The tyre dimension of 80/90-17 M/C 44P will be applied as stated in the project specification. The parameter for tread width, aspect radio and wheel diameter can be referred to Figure 3.15.

퐷𝑖푠푡푎푛푐푒 푡푟푎푣푒푙 = 푛푢푚푏푒푟 표푓 푐푦푐푙푒푠 푟표푡푎푡푒 푋 푐𝑖푟푐푢푚푓푒푟푒푛푐푒 표푓 푤ℎ푒푒푙 (3.1)

Figure 3.15 Tyre dimension index [42]

The verification of pneumatic pressure level inside the tubeless tyre is conducted in the experiment. This experiment consist of two parts which are to investigate the pressure level from atmospheric to project specification pressure level and pressure sensor reading verification via reverse engineering method. The pressure sensor measure the pressure level via potential difference. The linear relationship of pressure sensor reading in terms of voltage with respect to the pneumatic pressure reading in term of k Pa was obtained. The sensor capture the pneumatic pressure from atmospheric to project specification that is from 0 k Pa to 270 k Pa. The equation of straight line was obtained through the compiled sensor data with resolution of 15 k Pa.

The experiment was conducted to convert the sensor reading in term of voltage to kilopascal from the atmospheric level until project specification level. Another experiment was conducted to check the validity of equation of straight line. A new equation of straight line was generated with reverse engineering method. Comparison on both equations show that the trend line was identical with slope (m) 0.01 % and y-intercept (C) 15 % differences. The result was discussed and justified in Chapter 4. The real-time pressure condition was shown by the LCD with sampling time. The real- time pressure condition were categorized into over-inflation, under-inflation and ideal pressure condition. The over-inflation was defined as 20 % of pressure level higher than ideal pressure level

51 at 200 k Pa. Whereas, the uver-inflation was defined as 20 % of pressure level lower than ideal pressure level at 200 k Pa.

The experiment to study the received signal strength index for ideal system positioning was conducted. The experiment was carried out to determine the RSSI at receiver with multiple locations for transmitter at different angles with respect to receiver as shown in Figure 3.16. The data transmission take place at different angles when the wheel is rotating, hence it is important to check the signal strength from multiple angles. The transmitter was placed at every 90° to check on the RSSI. The RSSI was recorded with 360° that is four different transmitter surrounded the central receiver. The RSSI was tabulated in term of Mode instead of average to check on the highest occurrence of signal strength. The experiment was extended to determine the placement for central receiver. The central receiver was originally placed between front and rear transmitters with same displacement as shown in Figure 3.17. In order to differentiate the signal transmitted from which transmitter, the placement of central receiver was re-positioned. The re-position placement for receiver was verified via the RSSI captured at receiver from both front and rear transmitters. Another experiment was carried out to determine the RSSI at different rotation speed. The rotation speed was changed from slow to high, which is 21 to 100 rpm. This experiment’s purpose to check the signal strength received at huge dynamic range.

Figure 3.16 Multiples direction of transmitter

Figure 3.17 Side view of receiver placement

3.5 Chapter Summary

The new system is implemented with the combination of transmitter and receiver modules. The transmitter module captures the tyre condition with number of sensors, whereas the receiver module analyzed the captured data and provide a feedback system. The hardware is selected based on the comparison between its parameter that refer to research specification. The software involved in the research are purposed to program the system, ease the procedure for data transmission and data acquisition. The experiment setup for model of vehicle rotation behaviours is conducted by investigation of accelerometer’s behaviour and fixed the tyre dimension. The experiment setup for pressure sensor reading verification is conducted from the pressure level from atmospheric to project specification. The experiment result is verified with the reverse engineering method to ensure the accuracy of data acquisition. The experiment setup for ideal system placement is conducted by investigate on the RSSI at central receiver with transmitters at different displacement and with transmitter at multiple directions in term of degree. The placement of central receiver is determined through the investigation of RSSI from the transmitters at front and rear tyre.

CHAPTER 4

RESULTS AND DISCUSSIONS

4.1 Overview

The development of the tyre pressure monitoring system in term of tyre rotational behavior will be discussed in this chapter. Three major experiments were conducted, these include modelling of vehicle rotational behaviors such as acceleration, deceleration and sharp braking; pressure sensor readings verification via reverse engineering algorithm and the study of received signal strength index for ideal system positioning. Modelling of vehicle rotational behavior is done by analyzing the captured sensor data from rotating wheel. In addition, the sensor data was used in the estimation of vehicle travel distance. Besides that, another experiment was carried out to ensure the accuracy of pressure sensor analogue reading via reverse engineering method. Furthermore, the area of bluetooth signal coverage was tested based on several angles and position to find out the strongest RSSI at receiver to avoid any data lost due to distance length.

4.2 Modelling Of Vehicle Rotation Behaviour and Vehicle Distance Traveled Calculation

4.2.1 Accelerometer Performance Based On Wheel Rotation with Speed Variation

This experiment is conducted to identify the effect of the accelerometer at different rotational speed applied. The sensor data were stored in the raw log file that is generated by the terminal emulator program. Table 4.1 shows the tabulated result of three rotational axis with different rotational speed. The average of z-axis acceleration is not calculated in Table 4.1 because it fluctuates with the speed applied. Whereas, the x and y axis shows repeated sinusoidal amplitude under different speed applied. From the Table 4.1, it shows that y-axis will have negative sign when heading to minimum value, this is aligned with gravitational expansion theory,

푊푒𝑖푔ℎ푡 = 푚푎푠푠 푥 − (푔 푓표푟푐푒) (4.0)

The equation carries a negative sign change due to gravitational force as the weight in the downward direction. Theory explains this through physics as weight is not able to generate acceleration, but is the reactional force of it.

Table 4.1 Accelerometer result of different speed

Rotational Axis, m/s2

X axis Y axis Revolution , rpm Z axis Min. Max. Min. Max. Min. Max. point point point point point point 25 2.83 10.60 -12.56 11.66 42.09 62.62 50 4.85 8.91 -13.44 12.72 41.17 70.24 75 5.19 9.17 -13.54 13.09 45.34 75.57 100 3.55 11.15 -13.63 13.73 41.76 124.05 Average 4.11 9.96 -13.29 12.80 - -

Behaviors of roll, pitch, and yaw from the sensor were represented by the x, y, and z axis respectively as shown in Figure 4.1. Figure 4.1 shows the acceleration of multiple axis G-force over the sampled time. All the axis present a repeated sinusoidal waveform pattern and these are due to the constant rotational speed of vehicle wheel, which is 25 rpm.

The comparison of Figure 4.1 (25 rpm, low speed) and Figure 4.2 (50 rpm, high speed) shows that there are differences in the z-axis and y-axis whereas the x-axis remains the same throughout the overall experiment. It can be concluded that the z-axis and y-axis will be affected by different rotational speeds while x-axis is fixed. In Table 4.1 shows that the x-axis and y-axis will only fluctuate within the range of ≈ 4.11 to 9.96 for x-axis and -13.29 to 12.80 for the y-axis which represent tyre running at different rotation speeds with the constant tyre circumferences.

The pattern of waveform shows that the low speed experiment is performed with sinusoidal wave, whereas the high speed experiment is performed with triangular wave. Every single mark from the graph represented one sample data captured from the sensor. Triangular waveform is performed when the time required for the wheel to complete one cycle is shorter for high speed when compared to low speed. The sampling time was preset with 100 ms and remain constant throughout the experiments.

Z-axis was interpreted as acceleration of the vehicle by comparing the graphs in Figures 4.1 and 4.2 because it significantly increases from the range of ≈ 40 to 60 m / s2 to ≈ 50 to 70 m / s2 when the speed is increased from 25 rpm to 50 rpm. From low to high speed, the z-axis significantly shows the offset of ± 10 m / s2.

Based on the project specification and sensor placement inside the tyre, it can be concluded that the y-axis (pitch) can be used to measure the rotation of wheel since it is constant throughout different speeds whereas the z-axis (yaw) measures its acceleration. The sensor reading for x-axis and y-axis is repeated when the tyre runs in different rotational speeds with constant sensor facing direction.

Acceleration versus Time domain 70

2 50

Acceleration, m/s Acceleration, 0

1 6

36 11 16 21 26 31 41 46 51 56 61 66 71 76 81 86 91 96

111 106 116 121 126 131 136 141 146 151 -10 101

-20 Sampling time, sec

X-axis Y-axis Z-axis

Figure 4.1 The graph at 25 rounds per minute, rpm

Acceleration versus Time domain

60 2 50

Acceleration, m/s Acceleration, 0

1 6

56 96 11 16 21 26 31 36 41 46 51 61 66 71 76 81 86 91

106 111 116 121 126 131 136 141 146 151 -10 101

-20 Sampling time, sec

X-axis Y-axis Z-axis

Figure 4.2 The graph at 50 rounds per minute, rpm

4.2.2 Accelerometer Performance Based On Wheel during Braking / Stop Condition

This experiment is conducted to identify the effects of the accelerometer when a brake is applied to a running wheel. Table 4.2 shows the tabulated results of multiple numbers of vehicle brakes and stopped condition. There are a total of 5 vehicle brakes from different acceleration being recorded to reduce the inaccuracy. The experiment begins with tyre running at constant speed, continue with 5 times of sharp brakes after acceleration, and end up with fully stopped tyre at the 5th time of sharp brakes.

Table 4.2 z-axis behaviour during brake

Brake attempt z-axis, m/s2 Slope

140.86 − 43.33 Brake - 1 43.33 = −12.19 24 − 32 117.96 − 53.40 Brake - 2 53.40 = −21.52 48 − 51 115.37 − 44.09 Brake - 3 44.09 = −14.26 59 − 64 145.65 − 49.93 Brake - 4 49.93 = −15.95 68 − 74 114.90 − 48.57 Brake - 5 48.57 = −22.11 82 − 85 Full Stopped 50.25 (constant)

Average From Attempts 48.25

Standard Deviation 3.53

Figure 4.4 shows the waveform of accelerometer’s z-axis over the time taken. The green indicator represents acceleration whereas red indicator represents deceleration (sharp brake) from the graph, which supports the previous experiment 4.2.1 which stated the z-axis as wheel acceleration. When the wheel is increased in speed, the z-axis also increased and vice versa for opposite situation. The table 4.3 shows the time taken in second for different attempts at acceleration, deceleration and sharp brake condition.

The graph clearly stated five attempts of sharp brakes after quick acceleration. From the comparison of 5 attempts, it is discovered that the z-axis consistently drop to read less than 53.39 m / s2 with different deceleration magnitude and speed. The magnitude is the highest point of z- axis before the deceleration whereas the deceleration speed are calculated as slope as shown in Table 4.2. The values of z-axis and slope at ‘Brake-2’ is much bigger when compared to other attempts because it seems like none full stopped brake or slowing down instead of sharp brake behaviour.

At the sharp brake condition, the slope is obtained at a value that is less than ≈ -12 due to quick deceleration over time and thus affecting the accelerometer to perform 1 complete sinusoidal wave with full 360° rotation. The z-axis of accelerometer will significantly show sharp slope during braking condition.

Therefore, from the experiment, the behaviour of acceleration, deceleration and sharp brake can be estimated with two parameters which is the z-axis of accelerometer pointed at 48.24 (average) with standard deviation of 3.53 and having the slope less than -12.

Table 4.3 Description of tyre running condition

Rotation Behaviour Sampled time, second Constant (start) 1 to 17 Acceleration 18 to 21, 33 to 37, 51 to 54, 65 to 67, 75 to 77 Deceleration 24 to 32, 48 to 50, 59 to 64, 68 to 74, 82 to 85 Stopped (end) 86 to 101

Figure 4.4 Behaviour of z-axis at tyre acceleration, sharp brake and stopped condition

4.2.3 Accelerometer Performance Based On the Wheel Rotation Counter and Distance Travel Calculation

This experiment is conducted to obtain the number of cycles of the rotating wheel and distance travelled estimation based on the number of wheel rotations. The rotation speed begins with slow (21 rpm) then to fast (100 rpm) with each attempt to complete 30 cycles of rotation. The time taken from each test with different rotation speeds was tabulated in Table 4.4.

Table 4.4 Experiment results of rotation counter with different speed

Attempt Rotation speed, rpm Number of sample Time taken, sec per cycle to complete 50 cycles of rotation 1 21 29 2.9 x 50 = 145 2 25 24 2.4 x 50 = 120 3 46 13 1.3 x 50 = 65 4 100 6 0.6 x 50 = 30

Figure 4.6 shows the behaviour of 3 axis accelerometer at different rotation speeds with the same time taken. From the graph, it is discovered that the complete sinusoidal wave was formed with constant rotation speed as discussed in section 4.2.2 experiment. The sampled time for each sensor reading is equivalent to 0.1 sec or 100 millisecond, which means the gap between one sampled data to another sampled data requires 100 ms.

Based on the comparison from each graph with same target 50 cycles to complete, it is discovered that the higher the rotation speed, the lesser the number of sampled data and time taken. Apart from that, each single graph shows the consistency of data when the vehicle wheel is running at a constant speed. Throughout the different rotation speed, the maximum amplitude (both positive and negative) of y-axis remains the same, which is it fluctuates in between 0 to -20 m / s2 as discussed in section 4.2.1 experiment.

1 4 7

28 79 10 13 16 19 22 25 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 82 85 88 91 94 97

106 103 109 112 -20 100

-40 a) Acceleration vs Time (Test1)

2 Aceleroration, m/sAceleroration,

Sample time, sec X-axis Y-axis Z-axis

Figure 4.6 Acceleration versus Time graph based on test runs

The circumference of the wheel is fixed due to tyre size, therefore this parameter is able to be used for rotary calculation. The technique is shown in Figure 4.7. Since the y-axis will only fluctuate in a certain range with different rotation speeds, support and resistance lines were applied to analyze the y-axis reading. These two support and resistance straight lines are able to approbate as the range for verification as 1 cycle of rotation.

Tyre rotation with speed vary

2 15 + 6.5 support 10 5

1 5 9

13 97 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93

-5 17

141 101 105 109 113 117 121 125 129 133 137 -10

- 7.5-15 resistance Aceleroration, m/sAceleroration, -20 Sample time, sec X-axis Y-axis

Figure 4.7 Technique for wheel rotation identification

The number of rotations can be detected by using an array operator in programming language with input from support and resistance lines. For example, using an array to verify the sensor reading of y-axis as 1 cycle rotation: [+Y, -Y]. If the y-axis crosses over the support line as shown in Figure 4.7, the system will count in array as [1, 0], whereas when the y-axis crosses down the resistance line, system will count in array as [0, 1]. When the condition of [1, 1] which is the AND condition is fulfilled, this means that the tyre has rotated 1 cycle completely and stored in rotation counter at 1, before resets the array to the origin [0, 0]. Figure 4.8 shows the algorithm of rotation counter in term of coding for better explanation.

Figure 4.8 Algorithm for rotation calculation

The system will continuously check for the fulfillment of array and stack up the cycles of rotation counter. Tyre circumference is an important factor that controls the distance travelled by the vehicle, for example tyre with larger diameter performs longer displacement in 1 cycle, vice versa to tyres with small diameters. With tyre wear omitted, the circumference of tyre is able to determine by Equation 4.1 [46]:

2×푇푟푒푎푑 푊𝑖푑푡ℎ× 퐴푠푝푒푐푡 푅푎푡𝑖표 Tyre circumference = ([ ] + Wheel Diameter) 푥 휋 (4.1) 25.4

2× 80 × 0.90 Tyre circumference = ([ ] + 17) × 3.14159 = 71.21 𝑖푛푐ℎ푒푠 25.4

The circumference of the wheel is a constant variable through the distance travelled, with tyre wear omitted. The distance travelled by the vehicle can be estimated by the product of the total number of rotation and wheel circumference. Assuming the number of cycles rotate is 50 cycles through the equation (4.2) discussed, therefore the total distance travelled is [47] :

Distance travel = numbers of cycles rotate × circumference of wheel (4.2)

Distance travel = 50 cycles × 71.21 inches = 3560.5 inches ≈ 90.4367 meters

4.3 Monitoring On Pneumatic Pressure inside Tubeless Tyre

4.3.1 Investigate From Atmospheric Pressure Level to Project Specification

This experiment is conducted to identify the analogue pressure readings obtained by pressure sensor within the range from atmospheric pressure level to project specified pneumatic pressure level inside the tyre. The sensor reading will be tabulated gradually in descending order from the project specification level at 270 k Pa to atmospheric pressure level. The pressure sensor output reading is obtained by subtraction of Vout+ and Vout-. For one sample of data as shown in Figure 4.10 the absolute pressure reading obtained is 12 after subtraction, which is equivalent to a pressure reading at around 130 k pa.

Figure 4.10 Output reading show in terminal emulator program

Pressure Reading Obtained From Atmospheric Until Project Spec Level

20 y = 0.05142x + 4.9667

10 Absolute Absolute Pressure , volt 5

Pressure Gauge Reading, kpa

Figure 4.11 Histogram of Absolute Pressure versus Pressure Gauge Reading

Figure 4.11 shows the histogram that had plotted the absolute pressure against the pressure gauge. The histogram is plotted in kilopascal (kpa) instead of pounds per square inch (psi) due to better presentation in integers instead of decimal. Absolute pressure reading is obtained via pressure sensor, whereas pressure gauge reading is measured by the measurement tool. The histogram comes with standard error bar indicator due to the huge amount of sensor reading obtained and all the plotted values in the graph is obtained by average pressure reading among each pressure level. The evaluated sensor reading obtained linearly increases with pressure level.

The linear trend line shows that the absolute pressure gradually increase with pressure gauge reading with the resolution about 20 kpa from atmospheric pressure level until the project spec 270 kpa. By applying equation of straight line (4.3), it is able to form the equation for the trend line as shown in the graph.

y = mx + C (4.3)

Table 4.5 shows the differential voltage reading obtained via 100 samples with consider the absolute pressure and gauge pressure. The linear relationship between the differential voltage and pressure reading was determined.

Table 4.5 Differential Voltage reading respect to pressure reading Average Differential Average Voltage, volts Differential Pressure Reading Voltage, volts Standard (k Pa) (consider absolute Deviation pressure) (consider gauge pressure)

270 18.98 13.97 0.73 257 18.18 13.16 0.80 245 17.86 12.84 0.73 235 17.35 12.33 1.00 220 16.15 11.13 0.96 190 15.27 10.25 0.71 175 13.98 8.97 0.97 150 13.12 8.10 0.90 145 12.93 7.91 0.91 125 11.75 6.73 0.77 117 11.09 6.07 0.92 97 9.95 4.94 0.72 80 9.52 4.50 1.05 65 8.99 3.97 0.62 46 7.54 2.52 0.58 29 7.03 2.01 0.76 0 5.25 0.23 0.44 0 5.02 0 0.85

The calculation to form the equation is described as below: Consider two points from the pressure sensor voltage in Figure 4.11 histogram which are shown in Table 4.6.

Table 4.6 Parameter uses to calculate the equation of straight line

Pressure sensor voltage, Pressure gauge reading, Point y-axis x-axis

1st 18.1818 257

2nd 9.9545 97

Equation of a straight line, y = mx + C (4.4)

Given that, m = 0.05142 and C = 4.9667

After the calculations, the final equation was: y = 0.05142x + 4.9667. (4.5)

The actual pressure value that able to be calculated by the equation had been discussed in previous section, as shown in Figure 4.12:

Assumed pressure sensor voltage = 13 into Equation 4.5: 푦 = 0.05142푥 + 4.9667 13 = 0.05142푥 + 4.9667 13 − 4.9667 푥 = 0.05142 푥 = 156 푘푃푎

Figure 4.12 Gauge pressure reading at 156 kpa

Figure 4.12 shows the pressure gauge reading at about 156 kpa with the absolute pressure at 13 volts which is tally with the value calculated by the straight line equation. Based on the Table 4.5 the pressure reading has fluctuated lower than 1.00 according to the standard deviation, with this the tolerance of the system is obtained.

Due to pressure sensor limitations, the sensor reading only increments in integers, hence the resolution for the system can be assumed as 1. The tolerance and resolution is equal to 1, hence can be calculated in the same way in terms of kpa:

푦 = 1 + 5.0185 푦 = 0.05142푥 + 4.9667 6.0188 = 0.05142푥 + 4.9667 푥 = (|6.0188 − 4.9667|)/0.05142 푥 = 20 푘푃푎 = 2.90 푝푠𝑖

Where 5.0185 is the absolute reading at the atmospheric pressure. The standard atmospheric pressure is the pressure at any given point exerted by the earth's atmosphere and is the product of gravitational acceleration and the mass of atmospheric column of the unit area above the given point. The sensor reading used in this research is based on the absolute pressure reading. Absolute pressure is zero-referenced against a perfect vacuum, is equal to gauge pressure plus atmospheric pressure, therefore the overall atmospheric pressure is the addition of ambient pressure and tolerance. To further explain, gauge pressure is zero-referenced against ambient air pressure, so it is equal to absolute pressure minus atmospheric pressure with negative sign omitted. This calculation is proven with comparison of sensor data at atmospheric level and standard atmospheric pressure, which is 100.37 kpa and 101.3 kpa with minor offset caused by decimal during the calculation.

4.3.2 Pressure Sensor Reading Accuracy Test by Reverse Engineering

This experiment is conducted to ensure the accuracy of sensor readings obtained by pressure sensor via reverse engineering. The sensor reading will be tabulated from 180 kpa pressure level until 270 kpa pressure level (project spec) with 15 kpa of resolution as shown in Table 4.7.

Table 4.7 Pressure sensor readings by reverse engineering method Pressure pump Reference sensor Average Absolute Pressure Standard into tyre, kpa reading, volt reading from pressure Deviation sensor, volt 180 14 14.0912 0.5394 195 15 15.0912 0.3015 210 16 15.7272 0.4671 225 16 16.3637 0.6744 240 17 17.2727 0.4671 255 18 18.0909 0.3015 270 19 18.8181 0.4046

Figure 4.13 shows the graph of the actual pressure level inside the tyre against the sensor reading through reverse engineering technique. The y-axis is directly proportional to x-axis with a linear trend line obtained. The comparison between this experiment and previous experiment in section 4.3.1 was instead of measuring the pressure reading from project spec pressure level (270 k pa) to atmospheric pressure (0 k pa), this experiment pumps in specific pressure level into the tyre and predict on the sensor reading through calculation. The Table 4.7 shows that the absolute pressure reading obtained by pressure sensor is quite near to the predicted data which is ± 0. 3 when compared, which is within the project tolerance as discussed. The highest fluctuation range is 0.6744 at 225 k pa but still within the project tolerance.

Figure 4.13 Graph of Pre-set Pressure Reading versus Sensor Reading

The following calculation shows the reverse engineering procedure. The equation for the linear trend line was calculated with two points which is 255 k pa and 270 k pa:

Given that: m = 0.04848 C = 5.7285

After the calculations, the final equation was: y = 0.04848x + 5.7285 (4.6) With the comparison of equations from two pressure sensor experiments, although it shows the slightly different in slope. For percentage difference, m having 0.01% and C intercept point having approximately 15%. The slope percentage difference show almost identical linear trend line, whereas the intercept point shows a bigger difference due the reverse engineering method was calculated with preset pressure level when compared to experiment 4.3.1 which is random. Therefore it can be concluded that the prediction technique with reverse engineering is working fine, because the output is within project tolerance. The pressure level measured by the reverse engineering technique is aligned with experiment 4.3.1 and within the tolerance.

Equation 4.6: 푦 = 0.04848푥 + 5.7285 (reverse engineering method) Equation 4.5: 푦 = 0.05142푥 + 4.9667

The system for real-time tyre pressure measurement is shown in Figure 4.14. There was three possibility of tyre pressure conditions will shows by the LCD, such as ‘Over_inflat’, ‘Under_inflat’, and ‘Good_conP’ which represented the tyre over-inflation, under-inflation, and ideal pressure condition respectively. The real-time pressure condition shown in LCD is refreshed every 100 ms according to the pre-set sampling time.

Figure 4.14 Real-time pressure measurement on LCD

4.4 Data Transmission Quality on Bluetooth Connection

This experiment is conducted to determine the placement for the receiver module placement via the verification on received signal strength index (RSSI). The RSSI is affected by the displacement changed from transmitter to receiver. The RSSI is tabulated in Table 4.8 from multiples direction of the transmitter as illustrated in Chapter 3 Figure 3.16 with constant displacement. Since the RSSI drifts in a range because of surrounding environment influence, Mode has been used instead of the average of 100 numbers of sampled data which is shown in Table 4.8. Mode (most often happen) is able to reflect the actual signal strength and filter out the noise. The RSSI values are expressed as negative number due to the attenuation of free space and low power level. The range (min to max) of RSSI value depends on the hardware manufacturer. The RSSI remains constant at different directions of transmitter with the receiver is fixed at the center point among each transmitter’s position as illustrated in Figure 3.16. There is a total of 4 transmitter positions with 90° increment for each in order to complete one circular shape that covers the receiver with constant displacement.

Table 4.8 RSSI from multiples transmitter direction

Direction Angle, degree RSSI (mode) 0 ° -68 90 ° -69 180 ° -68 270 ° -67 360° -68

Based on the comparison from each transmitter point, the RSSI is consistent among four attempts with ± 1 of difference, means that the RSSI is changing with distance of transmission because of the power level being received by an antenna is changed. Therefore, the higher the RSSI number (nearer to zero), the stronger the signal is.

The transmitter sensor inside the front and rear tyre is able to identified by the RSSI captured. The original position of the central receiver is placed at the center point between the front and rear tyre with identical displacement as discussed in chapter 3 Figure 3.17. In order to differentiate the data transmit from different tyres, the central receiver is re-positioned (further from the rear tyre, while nearer to the front tyre) and the result was tabulated in Table 4.9. The RSSI value for rear tyre decreased after re-positioning when compared to the original receiver position due to the fact that the transmission distance is increased between the rear tyre and the receiver. Contrary, the value of RSSI for the front tyre increased after the central receiver re-positioned. The re-position placement for central receiver was shown in Figure 4.15 with position 1 and 5 as reference for rear and front tyre respectively. The position of central receiver was offset 10 cm from origin position. The result show that the RSSI after receiver re-positioned able to determine the location of transmitter (from rear or front) with the difference.

Table 4.9 Transmitter position in rear and front tyre

Original New RSSI after Original Position Displacement, Displacement, receiver re- positioning RSSI cm cm position 1 73 -72 83 -74 2 105 -78 115 -80 3 93 -76 103 -78 4 47 -68 57 -70 5 73 -72 63 -70 6 105 -78 95 -76 7 93 -76 83 -74 8 47 -68 37 -66

Figure 4.15 Re-position for central receiver

Table 4.10 shows the reading of RSSI with respect to the change of rotation speed, which is from slow to high speed respective to the attempt. In this table, the RSSI reading is tabulated in the form of average instead on mode due to the tyre being in a dynamic condition. From all the attempts, it is discovered that the RSSI reading shows consistency at reading around -71 with ±1, therefore can be concluded that the RSSI reading is not affected by changing in rotational speed. Hence this is aligned with theory stated that the RSSI is affected by displacement and determined by power level being received by an antenna. Therefore, the attempt of using RSSI to determine the quality of data transmission in several tyre rotation behaviour such as acceleration, de-acceleration and sharp brake condition is established.

Table 4.10 RSSI with different rotation speed Rotation speed, Attempt Average RSSI rpm

1 21 -70.67 2 25 -71.74 3 46 -71.16 4 100 -71.50

4.5 Chapter Summary

Three major experiments were conducted with each having the subsequence experiment. The major experiments included the modelling of vehicle rotation behaviour, pressure sensor reading verification and the ideal system positioning. There were three subsequence experiments conducted in the experiment of modelling of vehicle rotation behaviour. The vehicle rotation behaviour was investigate by the accelerometer performance based on different rotation speed applied. The result shown that the y-axis was constant throughout the different rotation speed, therefore it was use for wheel rotation measurement.

Contrary, the z-axis was fluctuated with different rotation speed, therefore it was use for vehicle acceleration measurement. The subsequence experiment follows with determine the z-axis performance when the wheel at acceleration, deceleration and braking/stop condition. The result shown that the acceleration, deceleration, braking/stop of the vehicle were identified with the magnitude and slope obtained from z-axis. The last subsequence experiment was vehicle distance travel measurement. The waveform pattern of wheel rotation was identified from y-axis by the system via array filtering process and from z-axis with multiple rotation speed applied. The distance travel by the vehicle was measured with the wheel rotary counter and constant tyre circumference.

The second major experiment was monitoring the pneumatic pressure inside the tubeless tyre. There were two subsequence experiments conducted. The first subsequence experiment investigated the pneumatic pressure from atmospheric pressure to project specification level. The

75 linear relationship between the differential voltage obtained pressure sensor and pressure reading was identified. The linear equation for pressure reading measurement that within project tolerance was generated from the result. The next subsequence experiment was conducted to ensure the accuracy of sensor reading obtained via reverse engineering method. Another linear equation was generated to compare with the linear equation generated in previous subsequence experiment. The result show that both equations were able to generated the same output. The third major experiment was conducted to determine the placement for central receiver with RSSI. The RSSI at receiver was tested under multiples transmitter direction and different rotation speed. The result show constant as RSSI will only change with displacement. The central receiver was re-position in order to differentiate the signal transmitted from different transmitter. The central receiver was placed in between front and rear transmitter with different displacement stated.

CHAPTER 5

CONCLUSION AND FUTURE WORKS

5.1 Conclusion

This research attempted to assess tyre rotation behaviour with tyre pressure monitoring system. There were three objectives that had achieved in the research, such as developed the system that calculate on the distance travelled by the vehicle, capable to capture the real-time tyre pressure condition in running tyre and the placement of central receiver was determined. The experiment of wheel rotation with speed variation is conducted to check on the system performance under different rotation speed and rotation behaviour such as acceleration, deceleration and braking/stop condition. The test showed that y-axis of accelerometer is always fluctuated within ±10 m / s2 with different rotation speeds and hence it is used to measure the number of cycles of rotated. Contrary, the z-axis of accelerometer was affected by the tyre rotation speed and hence the tyre rotation behaviour such as braking condition are capable to traced with the result having the slope less than -12. The result of experiments conducted provides the evidence support on the calculation of distance travelled by the vehicle via the product of numbers of cycles rotate and constant circumference of wheel.

Besides that, the experiment is conducted to capture the real-time tyre pressure condition in running tyre with accuracy ensured. The test showed that the differential voltage captured by the pressure sensor had the linear relationship respect to the pressure level inside the tyre. The pressure level was determined by the characteristics of differential voltage. The result are consistent with the standard deviation less than 1 from 100 sample data collected through the experiment. The result of reverse engineering method shows consistent with ±0.3 from the expected reading. The result of reverse engineering method shows 0.01% difference for slope (m) and 15% difference for y-intercept point (C) when compared, but the results obtained are within project tolerance of 1 psi. These results provide evidence support to the determination of pressure level. The final real- time tyre pressure condition are shown in the receiver module with LCD.

The placement of central receiver was determined with data transmission quality. The experiment is conducted with measurement of the RSSI at multiple transmitter directions and different tyre rotation speed. It was found that the RSSI of central receiver remains constant at different transmitter directions and different tyre rotation speed. The result are consistent with -68 with ±1 of difference under multiples transmitter direction and -71 with ±1 of difference under different tyre rotation speed. These results provide the evidence support to the assumption that RSSI affected by the change in displacement between the transmitter and the receiver. The ideal placement for central receiver was placed in between front tyre’s transmitter and rear tyre’s transmitter with is 73 cm in line with RSSI value at -72. In the case to differentiate the data transmitted from which transmitter, the placement of central receiver are moved 10 cm towards the front tyre’s transmitter. The final location for central receiver was placed in between both transmitters with 63 cm and 83 cm apart from front tyre’s transmitter and rear tyre’s transmitter respectively with RSSI value at -74 and -70.

5.2 Future Works

Specific conditions are fixed in order to carry out several experiments which are considered as a limitation of the research. For example, the vehicle heading direction is always constant and the placement of transmitter module needs to be faced up front and seated on the tyre rim. Further calibration is necessary on different models or types of vehicles due to the different sizing of tyres. A battery power source is also limited to a couple of hours for experiment data acquisition purposes. Low cost sensing devices limit the research sensitivity in term of efficiency, hence extra work around methods are necessary to ensure the sensor reading obtained.

Hence, these limitations are opportunities for future research, the tire rotation behavior can be further tested on a vehicle that runs on different pavement and slanted angles. Enhancements can be made by adding features, such as a GUI based automation calibration for all kinds of vehicle or tyre models. Further studies on power source may be required such as a sleep mode for power saving purposes or a self-powered transmitter module. The improvement of all hardware specs such as sensing devices is necessary in order to increase the system accuracy and efficiency for the development of self-driving vehicle or vehicle related applications.

REFERENCES

[1] MinistryTransportMalaysia, "General Road Accident Statistics in Malaysia (1997-2016)," ed: Department of Statistics Malaysia, 2016. [2] JKJR_Malaysia, "Statistic of Road Safety," 2016 ed: Ibu Pejabat, Jabatan Keselamatan Jalan Raya (JKJR) Malaysia, 2016. [3] DepartmentOfTransportation_UnitedState, "Critical Reasons for Crashes Investigated in the National Motor Vehicle Crash Causation Survey," ed, 2015. [4] C. Sharmila and V. Vibin Mammen, "Design of a real-time tyre pressure monitoring system for LMVs," in 2016 Online International Conference on Green Engineering and Technologies (IC-GET), 2016, pp. 1-4. [5] K. Seung Gi, C. Han Wool, L. Sol, and K. Dae Jin, "Design of open-ring monopole antenna for TPMS sensor modules," in 2016 Eighth International Conference on Ubiquitous and Future Networks (ICUFN), 2016, pp. 721-723. [6] Y. Wang, L. J. Wu, Z. Y. Tu, X. M. Zhang, and W. Jia, "A 125KHz low frequency power recovery circuit for battery-less TPMS SoC," in 2015 IEEE 11th International Conference on ASIC (ASICON), 2015, pp. 1-4. [7] A. Miclăuş, R. Jánó, and G. Chindriş, "Implementation of a car model for indirect tire pressure monitoring system," in 2016 39th International Spring Seminar on Electronics Technology (ISSE), 2016, pp. 316-321. [8] AVERAGE MONTHLY WEATHER IN GEORGE TOWN, MALAYSIA [Online]. Available: https://weather-and-climate.com/average-monthly-Rainfall-Temperature- Sunshine,george-town,Malaysia [9] (22th Feb 2017). Your guide to Tyre Pressure Monitoring Sytems (TPMS). Available: https://www.motoringassist.com/motoring-advice/car-maintenance/wheels-tyres-and- brakes/guide-tyre-pressure-monitoring-sytems-tpms/ [10] Bridgestone. WHAT IS TPMS & HOW DOES IT WORK. Available: http://www.bridgestonetire.com/tread-and-trend/drivers-ed/tire-pressure-monitoring- system-how-tpms-works [11] TIRERACK. (2017, 22 Feb). Tire Pressure Monitoring Systems (Direct vs. Indirect). Available: https://www.tirerack.com/tires/tiretech/techpage.jsp?techid=44 [12] IIIPERFORMANCE. (2017, 22 Feb). Things Should Know About TPMS. Available: http://www.pmwonline.com/things-know-tire-pressure-monitoring-systems-tpms/ [13] T. P. Systems, "Tire Pressure Systems," in Report of a study conducted by the North American Council for Freight Efficiency on the Confidence of Adopting Tire Pressure Systems, N. A. Council, Ed., ed. North American: North American Council, 2013. [14] S. N. Kadam, "A Review On Currently Available Different Self-Inflating Tyre Systems In Commercial Vehicles," presented at the International Journal of Advance Research, Ideas and Innovations in Technology., 2017. [15] B. Irvine. (2016, 25 Feb). What is the difference between radial and non-radial tires. Available: https://www.quora.com/What-is-the-difference-between-radial-and-non- radial-tires [16] S. Phillips. (2015, 23 Feb). What Tires Are Made Of20. Available: https://www.thoughtco.com/what-tires-are-made-of-3234478

[17] J. C.Glennon. (1998-2017, 25 Feb). Tire Failure Analysis. Available: http://www.crashforensics.com/tirefailures.cfm [18] Y. T. Sales. (2011, 25 Feb). Car Tyres and your Safety. Available: http://www.yorktyresales.co.uk/car-tyres-safety.html [19] T. lawyers. (25 Feb). Why Tires Fail [Tyre Failures]. Available: http://www.aa1car.com/library/tirefail.htm [20] CARTOQ. (2016, 25 Feb). Five points to consider when buying new tyres. Available: http://www.cartoq.com/five-points-to-consider-when-buying-new-tyres/ [21] RAC. (2016, 26 Feb). https://www.rac.co.uk/drive/advice/know-how/checking-tyre-tread/. Available: https://www.rac.co.uk/drive/advice/know-how/checking-tyre-tread/ [22] C. TIRE. (2017, 25 feb). Tire Dictionary. Available: http://www.coloradolandtire.com/tire-dictionary.html [23] Kioskea. (2014, 26 Feb). Bluetooth - How it works. Available: http://ccm.net/contents/69- bluetooth-how-it-works [24] A. Sable, "Comparative Study on IEEE Standard of WPAN 802.15.1/ 3/ 4 " International Journal for research in emerging science and technology, vol. 1, p. 28, 2014. [25] L. j. Chen, T. Sun, and M. Gerla, "Modelling Channel Conflict Probabilities between IEEE 802.15 based Wireless Personal Area Networks," in 2006 IEEE International Conference on Communications, 2006, pp. 343-348. [26] M. A. Haque and M. D. Hossain, "Technology survey of wireless communication for in- vehicle applications," in The 8th International Conference on Software, Knowledge, Information Management and Applications (SKIMA 2014), 2014, pp. 1-7. [27] Merriam. (2017, 26 Feb). radio frequency. Available: https://www.merriam- webster.com/dictionary/radio%20frequency [28] A. Dadashnialehi, A. Bab-Hadiashar, Z. Cao, and A. Kapoor, "Enhanced ABS for In- Wheel Electric Vehicles using data fusion," in 2013 IEEE Intelligent Vehicles Symposium (IV), 2013, pp. 702-707. [29] C. E. Tannoury, F. Plestan, S. Moussaoui, and G. P. Gil, "A variable structure observer for an on-line estimation of a tyre rolling resistance and effective radius," in 2012 12th International Workshop on Variable Structure Systems, 2012, pp. 167-172. [30] Newton. (26 Feb). Newton's Laws - Lesson 3 - Newton's Second Law of Motion. Available: http://www.physicsclassroom.com/class/newtlaws/Lesson-3/Newton-s-Second-Law [31] M. Islam and M. R. b. Bhuiyen, "Low cost digital stroboscope designed to measure speed of motor optically," in 2011 International Conference on Electronic Devices, Systems and Applications (ICEDSA), 2011, pp. 187-190. [32] A. M. Hendy, S. A. Hegazy, Y. H. e. Hendawy, and M. A. A. Emam, "Mechatronics system for tire pressure control," in 15th International Workshop on Research and Education in Mechatronics (REM), 2014, pp. 1-5. [33] R. K. Megalingam, C. Jayakrishnan, S. Nambiar, R. Mathews, V. Das, and P. Rao, "Automatic pressure maintenance system for tyres in automobiles to reduce accidents," in 2016 International Conference on VLSI Systems, Architectures, Technology and Applications (VLSI-SATA), 2016, pp. 1-6. [34] B. Shyrokau, D. Wang, L. Heidrich, and K. Höpping, "Analysis of subsystems coordination for electric vehicle during straight-line braking and brake-in-turn," in 2013 IEEE Symposium on Computational Intelligence for Engineering Solutions (CIES), 2013, pp. 61- 67.

[35] B. Chul Kyun, K. Seung Gi, L. Sol, and K. Dae Jin, "TPMS tire position registration algorithm using acceleration and RSSI data," in 2016 Eighth International Conference on Ubiquitous and Future Networks (ICUFN), 2016, pp. 745-747. [36] A. Sharma and N. Pandharpurkar, "The effect of wheel spin velocity on the propagation of signal to enhance the performance of tire pressure monitoring system," in 2015 International Conference on Industrial Instrumentation and Control (ICIC), 2015, pp. 1154-1157. [37] Element14. (2017, 27 Feb). Pressure Sensor Image. Available: http://au.element14.com/honeywell-s-c/nbpmpns060paunv/sensor-pressure-smt-low- 60psi/dp/2351998 [38] DealExtreme. (2017, 27 Feb). Inertial Measurement Unit image. Available: http://www.dx.com/p/gy-80-bmp085-9-axis-magnetic-acceleration-gyroscope-module- for-arduino-145912#.WSuuMWiGNEY [39] M. technologies. (2017, 27 Feb). Comparison of wireless technologies. Available: https://www.msc-technologies.eu/fileadmin/msc/images/business- unit/wireless/tables/_obsolete/table_ism_intro.jpg [40] SainSmart. (2017, 27 Feb). Wireless RF Transceiver Bluetooth Module HC-05 Master and Slave mode for Arduino. Available: https://www.sainsmart.com/wireless-rf-transceiver- bluetooth-module-hc-05-master-and-slave-mode-for-arduino.html [41] RobotShop. (27 Feb). Arduino Pro Mini 328 - 5V/16MHz. Available: http://www.robotshop.com/en/arduino-pro-mini-328-5v-16mhz.html [42] M. Tyre. (2017, 3rd March). Tyre Size Identification Guideliness. Available: http://motorcycle.michelin.in/tyre-basics [43] H. Fourati, N. Manamanni, A. B. Jemaa, L. Afilal, and Y. Handrich, "A quaternion-based Complementary Sliding Mode Observer for attitude estimation: Application in free- ranging animal motions," in 49th IEEE Conference on Decision and Control (CDC), 2010, pp. 5056-5061. [44] F. Edrisi, V. J. Majd, M. Attar, and N. Dini, "Modifying the attitude of quadruped robot body against disturbances via data fusion," in 2016 4th International Conference on Robotics and Mechatronics (ICROM), 2016, pp. 55-60. [45] S. Kuankid, T. Rattanawong, and A. Aurasopon, "Classification of the cattle's behaviours by using accelerometer data with simple behavioural technique," in Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2014 Asia-Pacific, 2014, pp. 1-4. [46] TIRERACK. (2017, 27 Feb). Calculating Approximate Tire Dimensions. Available: https://www.tirerack.com/tires/tiretech/techpage.jsp?techid=7 [47] MechanicTeacher. Wheels Diameter / Distance Traveled [Online]. Available: http://www.education.rec.ri.cmu.edu/previews/rcx_products/robotics_educator_workboo k/content/mech/pages/Diameter_Distance_TraveledTEACH.pdf [48] N. Baccar, M. Jridi, and R. Bouallegue, "Neuro-fuzzy localization in wireless sensor networks," in 2016 International Symposium on Signal, Image, Video and Communications (ISIVC), 2016, pp. 35-40. [49] O. Canovas, P. E. Lopez-de-Teruel, and A. Ruiz, "Detecting Indoor/Outdoor Places Using WiFi Signals and AdaBoost," IEEE Sensors Journal, vol. 17, pp. 1443-1453, 2017.

[50] J. P. Qin, Q. X. Deng, Y. T. Wu, H. B. Tong, and X. L. Su, "Indoor WSN Region Localization Based on Time Window Statistics," in 2016 International Symposium on System and Software Reliability (ISSSR), 2016, pp. 59-67. [51] T. M. Amirthalakshmi, P. Vijayakumari, K. H. S. Murugan, and M. Ayyadurai, "Design of SDCA using RSSI for mid-air collisions," in 2016 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), 2016, pp. 1-4. [52] H. K. Boyapati, R. K. Elubudi, S. Ungati, and M. Jain, "Implementation of RSSI indexed look up table based AGC for improved dynamic range of DSSS based wireless rf transceivers," in 2016 2nd International Conference on Next Generation Computing Technologies (NGCT), 2016, pp. 373-377.

APPENDIXES

1. Bluetooth Module

2. Pressure Sensor

3. Accelerometer (IMU)

4. Arduino Pro Mini