FACULTY OF ENGINEERING

DEPARTMENT OF MINING ENGINEERING

FINAL YEAR PROJECT
DESIGN OF AN ANT-COLLISION SYSTEM FOR
UNDERGROUND MINE HAUL TRUCK
CASE STUDY: GOLD MINE COMPANY (U) LIMITED
BY
MASETTE PRIESTONE
BU/UP/2020/2033
O775883151/0759324350
[email protected]
SUPERVISORS:
MR. LUBAALE SOLOMON AZARIUS
MR.MASERUKA BENDICTO

A final year project report submitted to the Department of Mining Engineering, Faculty of
Engineering and Technology in partial fulfillment of the requirement for the award of the
Bachelor of Science in Mining Engineering of Busitema University.
ABSTRACT
Accidents involving trucks in underground mines present a significant safety concern, often
resulting in injuries, fatalities, and operational disruptions. This project aims to address this
critical issue by designing an advanced anti-collision system tailored specifically for trucks
operating in underground mine environments. The primary objective is to minimize the risk of
collisions and enhance overall safety within the mine site.
This anti-collision system leverages a combination of cutting-edge technologies, including
ultrasonic sensor. By integrating these technologies, the system will provide real-time
monitoring and detection of potential collision hazards, such as other vehicles, stationary objects,
tunnel walls or personnel, within the mine's confined spaces.
Key features of the system include precise truck dynamic obstacle detection, and automatic
braking for collision avoidance strategies, audible warning system using a sound buzzer, display
warnings on the LCD and the LEDs, a communication module for sending of messages to
management for truck driver behaviors. Through continuous data analysis and predictive
modeling, the system will be capable of anticipating collision risks and initiating proactive
measures to prevent accidents before they occur.
The design process involves thorough consideration of the unique challenges posed by
underground mining environments, including limited visibility, uneven terrain, and restricted
maneuvering space. Human factors such as operator interface and user experience is carefully
integrated into the design to ensure ease of use and acceptance by truck operators.
DECLARATION
I Masette Priestone, hereby confirm that I authored this report and that it accurately reflects the
events and activities during the project development period. I assure that the research conducted
for this report was thorough, utilizing credible sources and suitable methodologies. Any external
sources utilized have been duly acknowledged, referenced, and credited.
This report has not been employed for obtaining any academic recognition by any individual or
educational institution.
Date…………………………………………………………….
Signature………………………………………………………
APPROVAL
I MASETTE PRIESTONE, submits this final year project report to the faculty of engineering
in the department of mining engineering for examination with approval of my supervisors.
Mr. LUBAALE SOLOMON AZARIOUS
Signature: …………………………………………………
Date: ……………………………………………………….
Mr. MASERUKA BENDICTO
Signature: …………………………………………………..
Date ………………………………………………………….
DEDICATION
I dedicate this report to my entire family for the tireless efforts to see me reach this level, may
the good lord reward you abundantly.
ACKNOWLEDGEMENT
I take this opportunity to thank the almighty GOD for granting me life, good health and the
ability to research and gather information that is integrated in this report.
I am deeply indebted to my dear supervisors: MR. LUBAALE SOLOMON AZARIUS and
MR. MASERUKA BENDICTO for the time, support, guidance, knowledge, and advice
provided to me during the preparation of this research proposal. May the almighty reward you
abundantly.
Contents
ABSTRACT....................................................................................................................................................2
DECLARATION..............................................................................................................................................3
APPROVAL...................................................................................................................................................4
DEDICATION................................................................................................................................................5
ACKNOWLEDGEMENT.................................................................................................................................6
1 CHAPTER ONE....................................................................................................................................11
1.1 Background................................................................................................................................11
1.2 Problem statement....................................................................................................................12
1.3 Objectives of study....................................................................................................................12
1.3.1 Main objective...................................................................................................................12
1.3.2 Specific objectives..............................................................................................................12
1.4 Justification................................................................................................................................13
1.5 Scope of study...........................................................................................................................13
1.5.1 Design scope......................................................................................................................13
1.5.2 Geographical scope............................................................................................................13
1.5.3 Time scope.........................................................................................................................13
2 Chapter Two......................................................................................................................................14
2.1 Introduction...............................................................................................................................14
2.2 MAIN CONCEPTS OF THE PROJECT............................................................................................14
2.3 COLLISION AVOIDANCE..............................................................................................................16
2.3.1 Forms of collision in underground mines...........................................................................16
2.3.2 Causes of truck collisions in underground mines...............................................................17
2.4 REVIEWS ON VEHICLE COLLISION AND ACCIDENT AVOIDANCE.................................................17
2.5 COMPARISION ON TECHNOLOGIES OF VARIOUS PAPERS USED TO AVOID COLLISION IN HAUL
TRUCKS..................................................................................................................................................19
2.6 Conclusion.................................................................................................................................21
2.7 Existing haul trucks used............................................................................................................21
2.7.1 Gaps identified within using this convectional haul trucks................................................21
2.8 Designed system........................................................................................................................22
3 Chapter 3...........................................................................................................................................23
3.1 Methodology.............................................................................................................................23
 3.1.1 Field studies.......................................................................................................................23
3.1.2 Tools used in data collection..............................................................................................23
3.1.3 Data collection...................................................................................................................23
3.2 Specific objective one: To determine the parameters contributing to collision accidents of haul
trucks 23
3.2.1 Data Collection...................................................................................................................24
3.2.2 Human Factors Analysis.....................................................................................................24
3.2.3 Data Analysis......................................................................................................................24
3.3 Methodology for objective two; Selecting and designing the system components...................25
3.3.1 System operation...............................................................................................................25
3.3.2 System functions................................................................................................................25
3.4 COMPONENT SELECTION...........................................................................................................25
3.4.1 Selection of the system components using Analytical Hierarchy Process (AHP)................26
3.4.2 Procedures applied to select the system components......................................................26
3.4.3 Key criteria for component selection.................................................................................27
3.5 Microcontroller selection...........................................................................................................27
3.5.1 Criterion weight table for microcontroller components....................................................27
3.6 Ultrasonic sensor selection........................................................................................................30
3.6.1 CRITERIA DETERMINATION................................................................................................30
3.6.2 NORMALISED PAIRWISE CRITERIA MATRIX........................................................................31
3.6.3 NORMALISED SENSOR PAIRWISE MATRIX.........................................................................31
3.7 Other components selected include..........................................................................................31
3.8 SYSTEM DESIGN.........................................................................................................................33
3.8.1 Hardware tools/Components............................................................................................33
3.8.2 Software tools....................................................................................................................34
3.9 System block diagram................................................................................................................34
3.9.1 The Control unit:................................................................................................................35
3.9.2 H Bridge:............................................................................................................................35
3.9.3 Warning Systems...............................................................................................................35
3.9.4 The Power Supply Unit.......................................................................................................36
3.9.5 Design and Interfacing for LEDs.........................................................................................36
3.9.6 GSM communication Module............................................................................................36
3.10 System design considerations....................................................................................................36
 3.10.1 FLOW CHART......................................................................................................................37
3.11 Automatic Braking System.........................................................................................................38
3.11.1 Braking system block diagram...........................................................................................38
3.11.2 Obstacle Detection System................................................................................................39
3.12 Modeling of a mathematical model for braking distance..........................................................39
3.12.1 ODE Equation:....................................................................................................................39
3.12.2 Relationship between vehicle speed, braking force and obstacle distance.......................40
3.12.3 Braking force applied on the brake pads by the servo motor............................................40
3.13 AUTOMATIC BRAKING SYSTEM DESIGN.....................................................................................40
3.13.1 System Design Assumption................................................................................................40
3.14 Methodology three: To test the performance of the system....................................................41
3.14.1 Unit Testing:.......................................................................................................................41
3.14.2 Integration Testing:............................................................................................................41
3.14.3 Full System Testing:...........................................................................................................42
4 CHAPTER THREE.................................................................................................................................43
4.1 RESULTS AND DISCUSSIONS.......................................................................................................43
4.2 Objective one result...................................................................................................................43
4.2.1 Data integration.................................................................................................................45
4.3 Objective two results.................................................................................................................46
4.3.1 Selection of the microcontroller........................................................................................47
4.4 Objective three..........................................................................................................................50
List of abbreviations

ACS Anti-Collision System

SDG Sustainable Development Goal

LiDAR Light Detection and Ranging
WSN wireless sensor network
LCD Liquid Crystal Display
IOT Internet Of Things
LEDs Light Emitting Diodes
SG Specific Gravity
AHP Analytical Hierarcy Process
SMS Short Message Service
Adx335 Accelerometer
L298N H-Bridge Motor driver
CPU Central Processing Unit
1 CHAPTER ONE
This chapter outlines the relevant information about the project, problem statement, and
justification, objectives of the study, purpose of the study and the scope of the study.
1.1 Background
Globally, the mining industry is a significant contributor to the economy, with haul trucks being
a crucial part of the mining process. However, haul truck collisions are a major safety concern,
resulting in injuries, fatalities, and significant financial losses (Kecojevic and Radomsky). The
increasing complexity and depth of underground mines, such as Wagagai Gold Mine, has led to a
higher risk of accidents, making it essential to implement effective safety measures (Rogers et
al., 2019).
Haul truck accidents are often caused by human error, inadequate safety measures, and limited
visibility(Zhang, Kecojevic and Komljenovic, 2020). The consequences of these accidents can be
severe, including loss of life, equipment damage, and production downtime. The economic costs
of haul truck accidents can be substantial(Bhateria and Jain, 2019). Moreover, the indirect costs,
including lost productivity and reputational damage, can be even higher.
Traditional safety measures, such as operator training and basic safety equipment, have
limitations in preventing haul truck accidents (Zhang, 2021). There is a growing demand for
advanced and automated safety systems that can detect potential collisions and alert operators in
real-time. Some advanced systems leverage technologies like GPS, radar, and LiDAR for real-
time monitoring, but their high costs make them less feasible for smaller mining operations.
In Uganda, haul truck accidents have occurred in several underground mines, including Wagagai
Gold Mine, resulting in injuries, fatalities, and production losses. The mine's haul truck fleet
operates in a confined and hazardous environment, with limited visibility and inadequate safety
measures, increasing the risk of accidents. The need for a cost-effective, real-time anti-collision
system that can continuously assess the risk of collisions and provide early warning alerts is
essential.
At Wagagai Gold Mine, the lack of an effective anti-collision system has resulted in several haul
truck accidents, putting the lives of miners and operators at risk. The mine's current safety
measures, including manual inspections and basic safety equipment, are inadequate to prevent
accidents. Therefore, there is an urgent need for a robust and cost-effective anti-collision system
that can detect potential collisions and alert operators in real-time, ensuring the safety of miners
and operators.
This project aims to design an Arduino-based Anti-Collision System (ACS) for haul trucks at
Wagagai Gold Mine, providing a cost-effective and robust solution to prevent accidents and
ensure the safety of miners and operators. The ACS will employ a network of sensors
strategically placed across the mine to detect potential collisions and alert operators in real-time,
minimizing the risk of accidents and ensuring the continuity of mining operations.
1.2 Problem statement
The high incidence of haul truck collisions in underground mines poses a significant threat to
miner safety and productivity. Despite existing safety measures, haul trucks are still involved in a
substantial number of accidents.
The primary causes of these collisions include limited visibility, inadequate communication, and
human error, exacerbated by the complex and hazardous environment of underground mines.
Conventional haul trucks lack real-time collision avoidance capabilities and often face
challenges in navigating through confined spaces and difficult terrains, which results into
accidents endangering both personnel and equipment( Patrucco et al. 2021).Lost productivity,
Increased maintenance costs, injuries, fatalities, production downtime and damage to equipment
and infrastructure( Dissertation of’, 2022) are realized. The operations of CASs utilizing manual
techniques such as visual surveillance, radio frequencies, and traffic control signs are affected by
confined spaces, dust and blast fumes and electromagnetic fields generated by electrical
equipment. This short coming conflicts with the United Nations Sustainable Development Goals
(SDGs) 9 and 12.( Giannetti et al. 2020)
Using a combination of sensors and real-time data analytics to detect potential collisions and
alert operators or automatically apply brakes to prevent accidents addresses the root causes of
haul truck collisions to improve mine safety, increase productivity, and minimize downtime,
ultimately enhancing the overall efficiency and sustainability of underground mining operations.

1.3 Objectives of study

1.3.1 Main objective
To design an anti-collision system for a haul truck.
1.3.2 Specific objectives
1. To determine the parameters contributing to collision accidents of haul trucks
2. To select and design the system components of the anti-collision system.
3. To test the performance of the system.
1.4 Justification
Haul trucks equipped with anti-collision systems will significantly be able to reduce the risk of
accidents and collisions and braking based on sensor data. This will lower repair costs for
damaged trucks and infrastructure, reduced downtime due to collisions translates to increased
operational uptime and improve worker safety.
1.5 Scope of study
1.5.1 Design scope
The system is designed using Proteus software to use ultrasonic sensors to detect collisions and
use linear actuators to automatically brake the truck in mine operations.

1.5.2 Geographical scope

This project is not limited to only Wagagai mining company as underground but can also be
applied to any surface mining company using convectional trucks to haul material..

1.5.3 Time scope

This project is limited to 3 months

2 Chapter Two
2.1 Introduction
This chapter includes a review of the technologies and the systems that are in existence to solve
the haulage related accidents elsewhere in the world, clearly showing their weakness and
pointing out the research gap that prompts our research, it includes performance of the proposed
system and states how it will solve the challenges of the existing system.

2.2 MAIN CONCEPTS OF THE PROJECT

Smart haul truck

A smart haul truck, also known as an autonomous haul truck or self-braking haul truck, is a large
vehicle used in mining operations for transporting materials such as ore, coal, or overburden
from one location to another within a mine site. The "smart" aspect refers to the integration of
advanced technologies that enable the truck to operate autonomously without human intervention
or with minimal human oversight. These technologies typically include; Sensors, Data
Processing, Artificial Intelligence (AI) and Machine Learning.( Sindi and Woodman, 2021)

A powered haulage accident

This occurs when a moving vehicle or equipment is involved in an incident that results in injury,
damage, or loss of life. This category includes accidents caused by the motion of the haulage unit
itself or failure of its components(Na and Hipertensiva,)

Collision of haul truck

Refers to accidents or incidents involving large vehicles specifically designed for transporting
materials such as ore, coal, or overburden within a mine.( Sindi and Woodman, 2021)

Anti-collision system

This is a system also known as a collision avoidance system technologically designed to prevent
or reduce the severity of collisions between trucks, pedestrians, or any other obstacle. It consists
of a combination of sensors, software ,warning and automatic braking systems that alerts the
driver to take control or takes over action to avoid an impending collision.

Automated system

Automated systems are designed to operate independently, combining sensors, controls, and
actuators to perform tasks with minimal or no human involvement. This is because humans are
susceptible to errors and fatigue, which can lead to various issues. Automated systems, on the
other hand, rely on integrated software and hardware that can function autonomously,
eliminating the need for human input and instruction for each operation, thereby enhancing
efficiency and reliability.
Microcontroller technology
The microcontroller refers to the use of microcontrollers which contains a processor, memory,
and programmable input and output peripherals and is designed for embedded applications. It is
the brain of the project, serving as the central component that governs the sensors monitoring the
garbage bin. Unlike microprocessors found in personal computers or other general-purpose
applications, microcontrollers are specifically designed for embedded applications. They are
ideal for use in automatically controlled devices and remote controls, making them a perfect fit
for this project. With their ability to process information and control peripheral devices,
microcontrollers enable the creation of intelligent, autonomous systems that can interact with
their environment. Microcontroller technology is used in a wide range of applications, from
simple consumer electronics devices, such as remote controls and toys, to more complex
applications such as automotive systems, industrial automation, and medical devices (Sachdeva,
2019)
GSM technology
GSM (Global System for Mobile Communication) is a widely used, open, and digital cellular
technology that enables mobile communication. It operates on four distinct frequency bands: 850
MHz, 900 MHz, 1800 MHz, and 1900 MHz. GSM combines Frequency Division Multiple
Access (FDMA) and Time Division Multiple Access (TDMA) technologies to facilitate efficient
communication. A GSM module is a compact chip or circuit that enables mobile devices or
computing machines to establish a communication link with the GSM network, allowing for
seamless connectivity and data exchange.
Sensing technology
Sensing technology is a cutting-edge innovation that utilizes sensors to gather information by
detecting and measuring physical, chemical, or biological properties, and converting them into
readable signals (Zhuang et al., 2023). A sensor is a device that detects and measures a specific
parameter, such as temperature, pressure, or light, and converts it into a usable output signal
(Ripka and Tipek, 2010).
Manual technology
Manual technologies used to detect collisions in a mine include visual inspections, audible
warnings such as horns or sirens, flagging systems to indicate equipment presence, radio
communication for coordination, regular safety meetings for education, hazard reporting for
corrective action, pre-operational checks on equipment, and traffic control measures like stop
signs and one-way signs to regulate movement. These manual technologies rely on human
observation, communication, and judgment, which can be prone to errors and limitations.
( Highhouse, 2008)
2.3 COLLISION AVOIDANCE
In relation to haul trucks it involves the implementation of technologies and strategies to prevent
or minimize the risk of collisions involving these large vehicles within mining operations( Sindi
and Woodman, 2021)

2.3.1 Forms of collision in underground mines

Side Collisions: Also known as T-bone collisions, side collisions occur when the side of one
vehicle is impacted by the front or rear of another vehicle traveling perpendicular to it.
( Chakraborty, Tsiotras and Diaz, 2013) These collisions occur at intersections or in confined
spaces where trucks are maneuvering.

Rollover Accidents: Rollover accidents involve a vehicle tipping over onto its side or roof.(Dr.
Vladimir) In underground mines, this happens due to sharp turns, uneven terrain, or high-speed
maneuvers, especially in larger vehicles such as haul trucks.

Fixed Object Collisions: Trucks can collide with stationary objects such as walls, support
structures, equipment, or debris within the mine( Naghshbandi, Varga and Hu, 2021). These
collisions often result from poor visibility, misjudgment of distance, or inadequate space for
maneuvering.

Pedestrian-Vehicle Collisions: In underground mines, pedestrians, including miners and

maintenance personnel, may be present alongside vehicle traffic.( Chehri et al., 2020) Collisions
between vehicles and pedestrians occur due to inadequate signage, poor communication, or
failure to observe safety protocols.

Multi-vehicle Collisions: In some cases, multiple vehicles may be involved in a collision, either
simultaneously or in a chain reaction. These incidents occur in congested areas, intersections, or
during emergency situations.

2.3.2 Causes of truck collisions in underground mines

Limited visibility, especially in large open-pit mines or underground tunnels, can leads to
collisions between haul trucks and other vehicles, equipment, or stationary objects.(Guenther and
Salow, 2015)
Operator Error i.e. Inattentiveness, fatigue, distraction, or misjudgment on the part of the haul
truck operator result in collisions. This might include failure to yield, improper lane changes, or
disregarding traffic signs and signals.

Equipment Failure such as brake failures, steering malfunctions, or tire blowouts can lead to loss
of control and collisions involving haul trucks( Bareket, Z., Blower, D. F., MacAdam, 2010).

Environmental Factors like adverse weather conditions such as rain, snow, fog, blast fumes, or
dust can impair visibility and road conditions, increasing the risk of collisions (Middleton, 2017)

Traffic Congestion or overcrowding in mining areas, especially at loading and dumping sites,
can increase the likelihood of collisions between haul trucks and other vehicles or equipment.

Human Factors like poor communication, lack of coordination between operators, and failure to
follow safety protocols can contribute to collisions involving haul trucks.( Reiman, 2020)

2.4 REVIEWS ON VEHICLE COLLISION AND ACCIDENT AVOIDANCE

The following literature survey is carried out with reference to vehicle collision and accident
avoidance
Adnan M. Al-Smadi, et al introduced a design aimed at detecting and mitigating potential
collisions, whether at the rear or front of a vehicle, to prevent or minimize the impact of such
incidents. The proposed method utilized an ultrasonic sensor, which senses when another vehicle
is approaching and alerts the driver accordingly. This system effectively gauges the distance
between two vehicles traveling in the same lane and direction. In cases where an object's
trajectory poses a danger to the vehicle, the system automatically activates safety measures to
protect the vehicle and its occupants (Abo-Hammour, Alsmadi and Al-Smadi, 2011).
The intelligent vehicle system enhancements are the foundation of this publication. Various units
that enhance the vehicle system are implemented in this text. Reducing medical support response
times and real-time accident detection are the primary objectives. The tire pressure is measured
to avoid accidents and accident detection is implemented using MCU nodes. MCU nodes are
used in the implementation of accident detection and prevention measures. MQ7 is employed in
pollution monitoring. The suggested technique lessens auto accidents, while pollution monitoring
contributes to environmental knowledge( Rakhonde, Khoje and Komati, 2018).
P. Ramya, the project goal was to offer a technical method for identifying and tracking driver
weariness so that collisions can be prevented before they happen to carry out. Finding out if the
driver is intoxicated was the goal. This locks the ignition system of the car. In order to prevent
accidents, this system regulates the direction of the car when the limit distance is crossed. We
also convey information to the appropriate owners or authorities in the event that the GSM
module was the cause of an accident. Based on observations, we may determine that driving-
related factors such as sleep, alcohol consumption, and overtaking are the three primary causes
of traffic accidents. The inability of some police officers to pull over every automobile and check
to see if the driver is intoxicated is one of the primary causes of drunk driving. As a result, an
efficient system for using alcohol detectors to screen for drunk drivers is required. The vehicle
may be detected moving ahead and controlled at a very high speed by connecting the Raspberry
Pi and the ultrasonic sensor (Murshed and Chowdhury, 2019). Alternatively, the vehicle can be
directed by turning left or right in accordance with the vehicle in another lane. An SMS will be
sent to the appropriate authorities or the car owner in the case of an unavoidable accident.
( Moniri-Morad et al. 2024), the project primary objective is to create a method for figuring out
alcohol content. When the driver's blood alcohol content surpasses the limit, the automobile
automatically cuts off the ignition and releases the air. Microcontrollers from the 8051 family
(89s52) are used in this project. This project uses MQ3 as an alcohol sensor to detect the amount
of alcohol in a person's breath. The 8051 micro-controller is unable to process the analog data
that the alcohol sensor produces. An analog-to-digital converter, also known as a digital
converter, transforms the data obtained from the alcohol sensor into digital format. After that, the
microcontroller stores the data and compares it to the threshold. The program controller will
regulate the ignition system if the value goes above the predetermined limit. Here, the ignition
system was managed by an electromechanical relay. In this project, it is feasible to prevent
accidents caused by driving while intoxicated by managing the ignition system
Mubashir Murshed numerous factors contribute to traffic accidents, with a significant portion
stemming from drivers' negligence and excessive speeds. Delays in reaching accident scenes due
to lack of awareness exacerbate the issue. However, the advent of Internet of Things (IoT)
technology presents a potential solution to mitigate accident rates. This article outlines an
intelligent system designed to detect and manage vehicle speed while promptly notifying
relevant parties in the event of an accident. Utilizing distance sensors, the system continually
monitors the gap between the vehicle and obstacles ahead. It intervenes by regulating speed and
alerting the driver to decelerate when the critical distance threshold is approached. In case of a
hazardous accident, the system automatically sends an email notification containing vehicle
details to the designated individual (Murshed and Chowdhury, 2019).
Firstly, there is an accelerometer circuit designed to detect drowsiness, and secondly, an alcohol
sensor circuit for identifying and controlling vehicle operation under the influence of alcohol.
The accelerometer circuit operates by sensing tilting motion, activating an alarm mechanism
through a relay to avert accidents caused by driver drowsiness. Similarly, the alcohol sensor
circuit triggers the relay to deactivate the vehicle's ignition process upon detecting alcohol levels
exceeding normal limits. Additionally, the system checks for brake failure and ensures
continuity, issuing warnings to the vehicle owner to prevent accidents resulting from brake
malfunctions.
The Mazda Algorithm suggests a method for estimating the optimal following distance required
between the lead vehicle and the host vehicle(Zhang, Antonsson and Grote, 2006)
The Berkeley Algorithm suggested determining the warning range Rw, which represents the
minimum buffer distance required to prevent collisions until both vehicles reach a complete
stop(Zhang, Antonsson and Grote). In contrast, the overriding range Ro, as calculated by the
Mazda algorithm, only accounts for the buffer distance needed from time 0 to τ. In this scenario,
the lead vehicle initiates braking at the maximum constant deceleration level −α, while the host
vehicle begins braking after a reaction time τ, also at the same deceleration level.
The Honda logic incorporates both warning and avoidance algorithms and presents two options
regarding whether the lead vehicle should stop before or after τ2 (Wang et al., 2016) The system
issues a warning when the range R falls below the warning range Rw and engages automatic
braking when R is less than Ro. In comparison to the Mazda algorithm, the Honda Algorithm
yields the smallest Ro.
2.5 COMPARISION ON TECHNOLOGIES OF VARIOUS PAPERS USED TO AVOID
COLLISION IN HAUL TRUCKS
Table 1.1( Siecker, Kusakana, and Numbi 2017)
Author Title Remark
Adnan M. Al-Smadi, Wasan “Intelligent In this article collision
Al-Ksasbeh, Mohammad Automobile Collision avoidance in rear and front of
Ababneh, Manar Avoidance and Safety Vehicles is only detected
Al-Nsairat (IEEE, 2020) System”. using ultrasonic sensor.
Obstacle detection or
collision avoidance is not the
only reason for accidents
occurrence.
Mahesh A. Rakhonde; Prof. ” Vehicle Collision Detection In this document pollution
Dr. S. A. and Avoidance with Pollution detection is done using MQ7
Khoje; Prof. R. D. Komati Monitoring System Using and accident detection which
(IEEE, 2018). IoT”. is not the only parameters
affecting the road safety,
hence further implementation
has to be done like driver
fatigue level, obstacle
 detection through LED lights
and displays alert message
through display.
P. Ramya, R.K. Kavi n, R. “Accident Avoidance In this article, only alcohol
Rathish, M. and Prevention detection and ultrasonic
Sathees Kumar, R. Karthi System”. sensors are implemented on
Kumar (IJESC, the
2020). Raspberry Pi, which slows
down
Communication between
vehicles a bit, which could
lead to an accident.
Aditi Padayar, “Microcontroller based This paper prevents IoT
Dipali Jadhav, Priti Accident Prevention System accidents, which are still slow
Pashte, Shweta Using IOT” communication for drivers.
Lagade, Prof. S. K Alcohol detection is the only
Srivastava (JETIR, parameter implemented in
FEB 2020). this paper.
Mubashir Murshed, “IoT-based Car Accident In this paper, the brake
Md Sanaullah Prevention and Detection detection parameters are
Chowdhury (ICATIS, 2019) System with Smart Brake implemented using IoT. As
Control” you know, the IoT deals with
the Internet. Slow
internet speeds can delay
email notifications and
vehicle details, leading to
accidents.

2.6 Conclusion
The conclusion drawn from the literature review underscores the critical importance of
implementing advanced collision avoidance systems, particularly in the context of designing and
developing an anti-collision system for trucks to enhance safe haulage. Given the emphasis on
factors such as driver fatigue detection, object detection, and effective communication among
vehicles, it became evident that incorporating smart technologies into truck design was
paramount for ensuring safe haulage operations.

Smart trucks equipped with state-of-the-art collision avoidance systems significantly enhances
safety in mines by actively detecting and mitigating potential hazards. For instance, a feature
such as automatic braking systems prevents collisions by swiftly responding to potential dangers,
thereby reducing the risk of accidents during transportation.

In essence, the conclusion drawn from the literature review underscored the urgent need to
design and develop an anti-collision system in haul trucks. By leveraging automation technology
and integrating features tailored to address key safety concerns, such as driver fatigue and object
detection, smart trucks can significantly enhance safety standards in haulage operations, thereby
safeguarding both drivers and mine workers.

2.7 Existing haul trucks used

Convectional haul trucks

2.7.1 Gaps identified within using this convectional haul trucks

Conventional haul trucks used in underground mining often lack advanced safety features
such as collision avoidance systems, emergency braking capabilities, or real-time monitoring
technologies, leaving them vulnerable to accidents.

Conventional haul trucks are typically designed for surface mining operations and may not
be adequately adapted to the unique challenges of underground environments.

Conventional haul trucks often operate independently without effective communication and
coordination mechanisms with other vehicles or central control systems.

2.8 Designed system

The designed system consists of an advanced Sensing Systems, Collision Avoidance Systems,
Emergency Response Mechanisms (automatic braking), and a Real-Time Monitoring system.

The cost incurred was only the initial costs to purchase the components and installation costs.
3 Chapter 3
3.1 Methodology
This chapter contains the research methods, tools and equipment, and procedures that were
followed to achieve the specific objectives towards the design and development of the real time
anti-collision system for Wagagai convectional haul trucks.
3.1.1 Field studies
This was achieved through several field visit at the company fleet maintenance department to
identify the parts and sides of the trucks with damages as a result of collisions. The damages on
trucks were identified using observation method under the guidance of the company safety
supervisor. This also involved several consultations from the truck drivers in obtaining data on
nature of collisions and how it affects the mine operations.

3.1.2 Tools used in data collection

Literature review databases like Google scholar and PubMed
Driver training records
Maintenance and safety inspection reports
Data loggers and sensors on haul trucks
Questionnaire approach

3.1.3 Data collection

During the primary data collection, the tunnel dimensions, the velocity of trucks, the distance
between the obstacles and the trucks were acquired.
The previous documented data on truck collisions accidents were analyzed and a graph of
variation to identify patterns and trends of their variation over time, identifying areas parts of the
trucks that was mostly affected by the collisions for sensor locations and this data guided me in
the programing of the sensors.

3.2 Specific objective one: To determine the parameters contributing to collision

accidents of haul trucks
This study employed a comprehensive mixed-methods approach methods (Almeida, 2018) to
assess the parameters contributing to collision accidents involving underground haul trucks.
This involved data acquisition, analysis and manipulation. Assessment of Current Infrastructure,
Traffic Patterns, Vehicle Characteristics, Human Factors, speed, distance, visibility, driver
fatigue and the obstacle detection, through literature review, data analysis, and expert interviews.
3.2.1 Data Collection
Literature Review: A thorough review of academic journals, government reports, and
industry publications was conducted to identify relevant factors contributing to collision
accidents.
Expert Interviews: Semi-structured interviews were conducted with 20 experienced haul
truck drivers and mine personnel to gather insights into safety culture, communication,
and human factors contributing to collisions.
 Survey Questionnaire: A survey questionnaire was administered to a representative
sample of underground mine drivers to gather information on their experiences, attitudes,
and behaviors related to safety and collisions.
Observations: Observations of drivers during their shifts were conducted to assess habits,
attention, and adherence to safety protocols.
Accident Reports and Incident Records: A thorough review of accident reports and
incident records was conducted to identify patterns and trends in collision accidents.
3.2.2 Human Factors Analysis
Driver State Assessment: Through review of the accident record accident book, I
analyzed the state of the human (driver) at the moment of the accident was investigated,
including fatigue levels, attention and distraction, emotional state, and physical condition.
Worker State Analysis: The worker's state at the time of the accident was examined,
including workload and stress levels, experience and training, and familiarity with the
mine environment.
Environmental Conditions: The environmental conditions that may have contributed to
the accident were assessed, including lighting system, dust levels and visibility,
temperature and humidity, and noise levels.
Accident Taxonomy: An accident taxonomy framework (HFACS) was applied to
categorize and analyze the human factors contributing to collisions.
3.2.3 Data Analysis
Descriptive statistics and frequency analysis were used to identify patterns and trends in the data.
I used linear regression analysis method to examine the relationships between collision accidents
over years to predict the trends for the next years
n ∑ XY −∑ X ∑Y
R= .
√ (n ∑ X −(∑ X ) )( n∑Y
2 2 2
−(∑Y )2 )

Thematic analysis was used to identify themes and patterns in the interview and survey data.
Content analysis was used to identify patterns and trends in the accident reports and incident
records.

3.3 Methodology for objective two; Selecting and designing the system components
System conceptualization
The Anti-Collision System (ACS) is designed to prevent collisions between haul trucks and other
vehicles, pedestrians, or obstacles in the mine. The system utilizes a combination of sensors,
processing units, and control systems to detect potential collisions and alert the driver or apply
the brakes to prevent an accident.
3.3.1 System operation
The system operates by continuously collecting sensor data on the surroundings, including
obstacle distance and truck speeds. This data is then analyzed by the processing unit to detect
potential collisions and determine the appropriate response. If a potential collision is detected,
the system alerts the driver through visual, audio warnings. If the driver fails to respond, the
system automatically applies the brakes to prevent the collision.
3.3.2 System functions
The functions of the ACS, include obstacle detection, collision prediction, warning systems, and
brake activation. The system detects obstacles, including other vehicles, mineworkers, and
environmental hazards, and predicts potential collisions based on sensor data. And the warning
system alerts the driver through multiple channels, and the brake activation feature is applied to
prevent a collision if the driver fails to respond.
The ACS is designed to reduce collision accidents, improve safety, increase efficiency, and
ensure compliance with industry safety standards and regulations.
3.4 COMPONENT SELECTION
The decision-making process regarding whether to make or buy the equipment involved
thorough research into existing designs and components. This included reviewing relevant
literature to understand the properties of available components in the market and comparing
them against the requirements of the necessary own components. Factors such as expertise levels
and incurred costs played a significant role in this decision-making process. Ultimately, the
chosen approach was to purchase pre-designed equipment that best matched the specified
requirements after careful consideration and evaluation. The microcontroller, sensors and
communication modules selection was based on the Analytical Hierarchy process (AHP).
This objective involves selecting and designing the specific components of the anti-collision
system, ensuring they are compatible, reliable and effective in underground mining environment.

3.4.1 Selection of the system components using Analytical Hierarchy Process (AHP)
This is a decision making tool used to select the best option from a set of alternatives based on
multiple criteria.
3.4.2 Procedures applied to select the system components
1. Defined the goal by determining the optimal system component.
2. Identified the criteria on which the system components will base on such as the system’s
Reliability, Cost, Performance, Compatibility and Maintenance.
3. Identified the alternatives for each component.
 4. Created a pairwise comparison matrix Comparing each criterion against every other
alternative using a scale of 1-9 where 1 is equal importance and 9 is extreme importance
A(Rahimdel et al., 2014) given below;

[ ]
1 a12 … a1 n
A= a21 1 … a2 n
a n1 an 2 … 1
5. Calculated the weight of each criterion and alternative using the Eigen value method
6. Calculated the score for each alternative by multiplying the weights by the ratings
(Rahimdel et al., 2014)

()
C1
T T
C= A . w = A . W n .1 = C 2 , i=1 , 2 ,3 … , n
Cn
Where A . wT follows,

( )
a 11 a12 a1 n
A . w = a21 a22 a2 n ∗( w1 , w2 ,.. w n ) T
T

an 1 a n2 a nn
7. Selected the alternative with the highest score for each component
8. Validated the results by checking for consistency and sensitivity analysis
The consistency value can be calculated using the following equation
CI
CV i= , i=1 ,2 , … n
wi

Maximal eigenvalue
n

∑ cv i
i
ἠmax = ,i=1 , 2 … n
n
Consistency Index
ἠmax −n
CI =
n−1
Consistency Ratio
CI
CR=
RI
Average Random Index
n−2
RI =1.98
n
3.4.2.1 Scale of relative importance of the attributes.
Scale Importance
1 Equal importance
3 Moderate importance
5 Strong importance
7 Very strong importance
9 Extreme importance
2, 4, 6 & 8 Intermediate values
1/3 , 5/2, 7/2, 9/2 Values of inverse comparison

3.4.3 Key criteria for component selection

a) Ensuring widespread availability and dependable sources.
b) Ensuring efficient and cost-effective fulfillment of requirements.
c) Ensuring the availability of essential software development tools such as compilers,
assemblers, and debuggers.
3.5 Microcontroller selection
Atmega 328p Arduino microcontroller was used. The choice of the controller was because of the
following reasons.
Easy to program with high level language programming using C, C++ and Java
It has relatively 13 digital and 5 analog pins, thus more inputs and outputs taken
Has 6 PWM channels, hence direct output into hazard and braking subsystems
Has relatively high memory capacity and inbuilt ADCs which makes it easy to interface with
analog inputs band on board encoders.

3.5.1 Criterion weight table for microcontroller components

When selecting a microcontroller, its cost is of extreme very strong importance, the user friendly
is of strong importance and the power consumption is of moderate importance

Different microcontrollers
Criteria Weight Arduino Uno Arduino Mega Raspberry PI
Cost 0.5 7*0.5 3.5 9*0.5 4.5 5*0.5 2.5
User friendly 0.3 5*0.3 1.5 7*0.3 2.1 3*0.3 0.9
Power 0.2 2*0.2 0.4 2*0.2 0.4 1*0.2 0.2
consumption
Sum 1.0 5.4 7 3.6
A pairwise comparison matrix Comparing each criterion against every other

Criteria 0.5 0.3 0.2

C1 C2 C3
Cost 1 0.6 0.4
user friendly 1.666666667 1 0.666666667
power consumption 2.5 1.5 1
Sum 5.166666667 3.1 2.066666667

Normalised pairwise matrix of the criterias

C1 C2 C3 Criteria weight
C1 0.193548387 0.193548387 0.193548387 0.193548387
C2 0.322580645 0.322580645 0.322580645 0.322580645
C3 0.483870968 0.483870968 0.483870968 0.483870968

COMPAIRING COSTS AGAINTS OTHER CRITERIAS

3.5 4.5 2.5

ARDUINO UNO ARDUINO MEGA Rasperry PI
ARDUINO UNO 1 1.285714286 0.71428571
ARDUINO MEGA 0.777777778 1 0.55555556
Rasperry PI 1.4 1.8 1
3.177777778 4.085714286 2.26984127

NORMALIZED COST PAIRWISE MATRIX

ARDUINO UNO ARDUINO MEGA Rasperry PI Criteria weight

ARDUINO UNO 0.314685315 0.314685315 0.31468531 0.314685315
ARDUINO MEGA 0.244755245 0.244755245 0.24475524 0.244755245
Rasperry PI 0.440559441 0.440559441 0.44055944 0.440559441

CHECKING CONSISTENCY OF THE COSTS

ARDUINO Criteria
UNO ARDUINO MEGA Rasperry PI weight
ARDUINO UNO 0.314685315 0.404595405 0.224775225 0.944055944 3
ARDUINO MEGA 0.19036519 0.244755245 0.135975136 0.571095571 2.333333
 Rasperry PI 0.616783217 0.793006993 0.440559441 1.85034965 4.2
λmax 3.177778
CI 1.677778
CR 1.89404
RI 0.66

COMPAIRING USERFRIENDLY
PAIRWISE MATRIX

1.5 2.1 0.9

ARDUINO UNO ARDUINO MEGA Rasperry PI
ARDUINO UNO 1 1.4 0.6
ARDUINO MEGA 0.714285714 1 0.428571429
Rasperry PI 1.666666667 2.333333333 1
3.380952381 4.733333333 2.028571429

NORMALIZED PAIRWISE MATRIX

Criteria
ARDUINO UNO ARDUINO MEGA Raspberry PI weight
ARDUINO UNO 0.295774648 0.295774648 0.295774648 0.295774648
ARDUINO MEGA 0.211267606 0.211267606 0.211267606 0.211267606
Rasperry PI 0.492957746 0.492957746 0.492957746 0.492957746

COMPARION OF WEIGHTED WEIEGHTS

weighted ARDUINO
average UNO Raspberry PI ARDUINO MEGA
Cost 0.193548387 0.060906835 0.077020881 0.107829234
user friendly 0.322580645 0.095411161 0.062487592 0.104146003
power consptn 0.483870968 0.120967742 0.0625 0.125
sum 0.277285738 0.202008473 0.336975237

3.6 Ultrasonic sensor selection

The HCSR04 ultrasonic sensor modules are selected in this project to determine the accuracy of
obstacle distance using the principles of sound wave propagation and echo detection. It consists
of a transmitter and a receiver, the transmitter emits high-frequency sound waves (ultrasonic
pulses) towards the target area and the emitted sound waves propagate through the air at a known
speed, usually at the speed of sound and travel outward in all directions from the sensor's
transmitter.
Ultrasonic sensor measures up to a maximum distance of 400cm and a minimum distance of
3cm.
Wire connection for the ultrasonic sensor modules is described as:
VCC is connected to 5V Supply.
Trig, which is Trigger Pulse Input is connected to microcontroller digital input/output pin.
Echo which is Echo Pulse Output is connected to microcontroller digital input/output pin.
GND is connected to 0V Ground.

Different microcontrollers

Criteria Weight HC-SR04 JSN-SR04 ultrasonic MB1010 Ultrasonic

Ultrasonic sensor sensor
sensor
Accuracy 0.4 10*0.4 4 8*0.4 3.2 7*0.4 2.8

User friendly 0.3 8*0.3 2.4 5*0.3 1.5 3*0.3 0.9

Cost 0.2 10*0.2 2 7*0.2 1.4 5*0.2 1

Power 0.1 6*0.1 0.6 9*0.1 0.9 8*0.1 0.8

consumption
Sum 1.0 9 7 5.5

3.6.1 CRITERIA DETERMINATION

0.4 0.3 0.2 0.1
USER
ACCURANCEY FRIENDLY COST POWER CONSUMPTION
ACCURANCEY 1 0.75 0.5 0.25
USER FRIENDLY 1.333333333 1 0.666666667 0.333333333
COST 2 1.5 1 0.5
POWER
CONSUMPTION 4 3 2 1
8.333333333 6.25 4.166666667 2.083333333
3.6.2 NORMALISED PAIRWISE CRITERIA MATRIX
POWER WEIGHTED
ACCURANCEY USER FRIENDLY COST CONSUMPTION AVERAGE
ACCURANCEY 0.12 0.12 0.12 0.12 0.12
USER FRIENDLY 0.16 0.16 0.16 0.16 0.16
COST 0.24 0.24 0.24 0.24 0.24
POWER
CONSUMPTION 0.48 0.48 0.48 0.48 0.48
3.6.3 NORMALISED SENSOR PAIRWISE MATRIX
HC-SR04
Ultrasonic JSN-SR04 MB1010 Criteria
sensor ultrasonic sensor Ultrasonic sensor weight
HC-SR04 Ultrasonic sensor 0.27184466 0.27184466 0.27184466 0.27184466
JSN-SR04 ultrasonic sensor 0.339805825 0.339805825 0.339805825 0.339805825
MB1010 Ultrasonic sensor 0.388349515 0.388349515 0.388349515 0.388349515

MB1010
WEIGHTED Ultrasonic JSN-SR04 HC-SR04 Ultrasonic
AVERAGE sensor ultrasonic sensor sensor
ACCURANCEY 0.12 0.032621359 0.092374392 0.131963418
USER FRIENDLY 0.16 0.030379747 0.057683064 0.153821503
COST 0.24 0.054193548 0.072840791 0.145681581
POWER
CONSUMPTION 0.48 0.198620689 0.114149822 0.085612366
SUM 0.315815344 0.337048068 0.517078868

When selecting an ultrasonic sensor, its accuracy is of extreme very strong importance, the user
friendly is of strong importance, the cost is of moderate importance and the power consumption
is of importance.
3.7 Other components selected include
16x2 LCD module with 12C module
Liquid crystal display is a type of flat panel display that uses liquid crystals to produce images.
This 16x2 12c LCD module is made for the Arduino microcontroller. It uses the 12C
communication interface to display information on the Arduino board. It is able to display 16x2
characters on 2 lines. They are inexpensive, simply programmable.

Figure 1;16x2 LCD module 12C module

Jumper wires
A jump wire is an electrical wire or group of them in a cable, with a connector or pin at each end
which is normally used to interconnect the components of a breadboard or other prototype or test
circuit internally or with other equipment or components, without soldering
Figure 2; jumper wires

Rechargeable lithium ion batteries

This offer high energy density, long cycle life and low self- discharge rate. They have a nominal
voltage of 3.6-3.7v. They are widely used in portable electronics

Figure 3; 3.7v rechargeable li-ion battery

Active buzzer
They can produce loud, high pitched sound thus serving as an audible warning. They have the
built in oscillator and do not require an external oscillating circuit. They are powerd by DC
voltage. They are polarized and designed to be compact and durable

Figure 4; active buzzer

LEDs
This displays real time system status through color- coded lights. They are energy efficient, long
lasting and environmentally friendly. They are highly durable and resistant to shock, vibrations
and extreme temperatures
Figure 5: LEDs

3.8 SYSTEM DESIGN

To achieve successful design and development of the system, it involved undertaking several
processes and the usage of various development tools such as hardware and software tools.

3.8.1 Hardware tools/Components

To design the hardware components of the system prototype, the following tools/resources were
used.

Laptop: This device was used to provide interface through which programs were to be
developed.

Microcontroller: This is a chip that held Arduino programs used to control the embedded
system.

LCD: this displayed an emergency message for the driver alerting him to take action prior to
the obstacle detected to mitigate collision.

Bluetooth module: this enabled me to control the truck using a smart phone

Motor driver: this enabled the connection of the dc motors onto the Arduino board

Impact sensors: these were used to determine the magnitude on impact exerted on a vehicle.

Jumper wires: these were used to connect different components in the system.

Breadboard: this where all the connections were done.

Servo motors; this were used to demonstrate the automatic braking system

LEDs; this were used to signal the driver of the obstacle ahead

Buzzer; this were used to issue an audible warning alert signal to the driver.
3.8.2 Software tools
Proteus IDE: this software was used for simulation purposes.
Arduino IDE: This is environment through which programs were developed from that it
was controlling the system.
MATLab software: used to generate the code used run the mathematical model of a
linear braking system.
MIT App Inventor: For designing application that controlled embedded systems
through Bluetooth.

3.9 System block diagram

The system comprises of three main parts ie the obstacle detection , warning and display system
and the automatic braking system with the following components installed;
1. Ultrasonic sensors installed on the front, rear and the sides of the truck to detect
obstacles at close range
2. Navigation and control system like the Bluetooth module that integrates sensor data to
control the movements of the truck.
3. Warning and alert system which include the audible warning system installed on the
truck cabin and the visual warning system (LEDs and LCD) installed on the dash board
and exterior of the truck
4. The automatic braking system (linear motion actuators) installed on the truck to slow and
stop the truck prior to the distance to collision.
 5. Power and communication system (batteries or power sources) installed to supply energy
to the truck, wiring and antenna installed to connect and communicate between
components.
3.9.1 The Control unit:
The control subsystem consists of the Arduino circuit controlled by a program loaded in its
memory. The Arduino circuitry consists of an Atmega 328p microcontroller chip, a 16 MHz
crystal oscillator. The 16 MHz crystal is used for timing; the reset button is used to reset the
microcontroller to upload a new program or to start running the existing program from the
beginning.

3.9.2 H Bridge:
Is an electronic circuit which enables voltage to be applied across a load in either direction,
where h is often used in various applications to allow DC motors to run in forward and backward
motion.
3.9.3 Warning Systems
This subsystem provides reliable warning to drivers about a possible collision prior to their
reaching an unsafe position or location. The composition of the warning system comprises of the
buzzer, LED and LCD. The mode of operation of the warning system is, a red LED will light
upon detecting an object, if the object is 3 meters away while being displayed on the LCD, the
buzzer starts buzzing, when the object is 2 meters close to the automobile the Precautionary
braking system to avoid collision is engaged.
The buzzer system for the anti-collision system consists of multiple buzzers with different
frequencies and sound patterns to indicate varying levels of collision risk. The buzzer is designed
to produce a gradual warning system, with a low-pitched beep (500 Hz) for minor obstacles, a
medium-pitched beep (1 kHz) for moderate obstacles, and a high-pitched beep (2 kHz) for severe
obstacles or short-distance warnings.
The buzzer is installed in strategic locations inside the truck, such as the dashboard, steering
column, or door panels, to ensure clear audibility for the driver. The buzzer system provides a
simple yet effective way to alert the driver of potential collisions, allowing them to take
appropriate action to avoid accidents
3.9.4 The Power Supply Unit
The power supply system designed in this project supplies the desired voltages to the micro-
controller and the buzzer, LCD and LEDs. The system operates at a voltage of 5V DC.
Therefore, from a voltage source of (7-12V) DC in this case, 12V battery will be used to power
the prototype, a voltage regulator is used to give stable 5 volts DC, however, the regulated output
usually varies between 4.8-5.2V.
3.9.5 Design and Interfacing for LEDs
LEDs are operated from a low voltage DC supply, with a series resistor to limit the forward
current to a suitable value of about 5-6Ma
3.9.6 GSM communication Module
The communication system design for the anti-collision system utilizes a GSM module to enable
wireless communication between the truck and a central server or emergency services. The GSM
module is integrated with a microcontroller and a SIM card, allowing it to send and receive SMS
messages. In the event of a potential collision, the system sends an automatic alert to the central
server or emergency services, including location data and collision details. The GSM module
also enables remote monitoring and configuration of the anti-collision system, ensuring real-time
updates and optimal performance. This design provides a reliable and efficient communication
system for the anti-collision system, enabling swift response and action in emergency situation
3.10 System design considerations
The design considerations of this system is that the ultrasonic sensor models will be designed to
measure and detect the obstacle and measures the distance prior to collision, data processing will
be based on Arduino board models to process sensor data and apply algorithms to detect the
obstacles and measure the distance. The motor driver will be designed to simulate and control the
dc motor based on sensor data. The braking system will include the linear actuators designed to
simulate the braking system of the truck. The display unit will be designed to simulate the
display of distances and warning messages and the braking algorithm will be developed to
determine the optimal braking force based on sensor data.

3.10.1 FLOW CHART

3.11 Automatic Braking System
3.11.1 Braking system block diagram

A linear actuator is an actuator which creates motion in a straight line, in contrast to the
circular motion of a conventional electric motor. Linear actuator is used in this system for
generating motion and automatic brake. A servo motor is a high-precision rotary actuator that
offers precise control over angular position, acceleration, and velocity. It builds upon the
capabilities of a regular motor, enabling specialized operations. For the prototype, the Servo
Motor SG90 was utilized to automatically brake the system, but a more powerful servo motor
will be required to integrate with the car's mechanism. Fortunately, the configuration remains
consistent across servo motors, making it easy to upgrade without requiring significant
changes.
3.11.2 Obstacle Detection System
Real-Time Distance Measurement
To model an effective obstacle detection system in an underground mine tunnel, is by employing
a real-time distance measurement approach using ultrasonic sensors. The time taken for the
ultrasonic waves to travel to and from the obstacle was multiplied by 2 to account for the to-and-
fro movement, and the real-time distance (d) is calculated using the equation
Total distance travelled by the ultrasonic wave∗speed of sound
Obs( d)=
2

3.12 Modeling of a mathematical model for braking distance

Model Assumptions:
Assumptions:

Constant braking force F b

Negligible air resistance and rolling resistance

Initial velocity Vi and initial kinetic energy KEi
Variables:
d is braking distance at time t
v ( t ) is Velocity at time t

KE (t): kinetic energy at time t

F b is constant braking force

m: mass of the truck

t: time
3.12.1 ODE Equation:
t

d
=v ( t ) … … … … … … … … … … … … … … … … … … … … … … … … …1
dt

dv −F b
= ……………………………………… ……………………………2
dt M
d (KE)
=−V ( t ) F b … … … … … … … … … … … … … … … … … … … … … … . … 3
dt
Initial Conditions:
d(i) = 0 (initial braking distance is 0)
v(i) = Vo(initial velocity)
KE(i) = KEi (initial kinetic energy)
3.12.2 Relationship between vehicle speed, braking force and obstacle distance
The relationship of the truck speed, braking force and obstacle distance was determined by
running the mathematical model in Matlab to create a realization of the model.
The MATlab code was generated in the MATlab environment to test the performace of the
model.
3.12.3 Braking force applied on the brake pads by the servo motor
To determine the braking force applied on the brake pads, the relationship between the braking
force and the force applied on the brake pads was related using the principle of mechanical
advantage (MA) which is designed to amplify the input force to produce a greater output force
that counter balances the braking force of the whole truck bring the truck to a stop.
The mechanical advantage is represented by the following equation
Braking force
MA=
Brake pad forc e
The mechanical advantage refers to the amplification of force achieved by the brake components.
It depends on factors such as brake pedal ratio, master cylinder ratio, and brake caliper ratio.
The mechanical advantage of the brake system depends on the specific design and configuration
of the brake components, including the brake pedal ratio, master cylinder ratio, and brake caliper
ratio.
Brake Pedal Ratio: The brake pedal ratio determines the force applied by your foot on the brake
pedal. A higher ratio amplifies the force transmitted to the master cylinder.
Master Cylinder Ratio: Converts pedal force into hydraulic pressure. A larger master cylinder
bore provides more force multiplication.
Brake Caliper Ratio: The brake caliper ratio affects the force applied to the brake pads. A larger
caliper piston diameter increases braking force.
3.13 AUTOMATIC BRAKING SYSTEM DESIGN
The automatic braking system using linear actuators is designed to enhance safety in
underground mine operations with haul trucks. This innovative system utilizes linear actuators to
automatically apply the brakes in emergency situations. The actuators convert electrical energy
into linear motion, providing precise and reliable control over the braking system. In this system
linear actuators are integrated with sensors and control units to detect potential collisions and
automatically apply the brakes to prevent the impact.
3.13.1 Mechanical Design of the modified servo braking system
The mechanical design of the servo motor braking system involved modifying the existing
manual braking system to accommodate the servo motor's pull action. The design focused on
ensuring a smooth and linear motion, minimal interference with the existing system, and optimal
performance.
Brake Pedal Modification
A custom bracket was designed to attach the servo output shaft to the brake pedal, ensuring
precise alignment and secure attachment. The brake pedal was modified to accommodate the
servo motor's pull action, providing a push force on the clutch and brake.
Servo Output Shaft Modification
A custom output shaft adapter was designed to connect the servo motor’s shaft to clutch and
brake using the clutch wires, ensuring secure attachment and proper alignment.
Clutch Wire Integration
Existing clutch wires were utilized to transmit the servo motor's pull action to the clutch pedal
and the brake pedal, ensuring proper tension and secure attachment.
Mounting Brackets
Custom mounting brackets were designed to securely attach the servo motor to the existing
manual braking system, ensuring precise alignment and secure attachment.

3.13.2 Relating brake pad force with the servo motor shaft deflect

Maximum overall braking force is F max

Total motor deflection (θmax ¿ 180 degrees (or π radians)

Brake pull force Fvaries from 0 to F max

Motor deflection θ varies from 0 toθmax

A linear relationship between the brake pull force (F) and motor deflection (θ ¿) as follows:
F max
F= θ max
θ

This equation indicates that the brake pull force (F) is directly proportional to the motor
deflection θ
The safe obstacle braking distance is 10m
The truck travels at a constant speed of 30km/hr which is the speed limit at Wagagai gold mine.
The mechanical advantage of the braking system of Wagagai haul trucks is 16 considering the
braking pedal ratio; brake caliper ratio and master cylinder ratio of 4:1, 2:1, and 2:1 respectively
The braking force is determined from the relationship graph of braking force and velocity
generated using MATlab software.

3.14 Methodology three:

To test the performance of the system.
The testing protocols followed three phases namely;

Unit testing
Integral testing
Full system testing

3.14.1 Unit Testing:

This was to verify individual components (units) of the system.

Procedure:

Isolated each module or function within the anti-collision system.

Created test cases to validate their correctness by uploading a single code at a time
Executed the tests and ensured that each unit behaves as expected.

Metrics:

Measured the percentage of code executed during testing.

Assess the reliability of individual units in different environment scenarios.
Evaluated the efficiency of the single components.

3.14.2 Integration Testing:

This validated interactions between different components.

Procedure:

Combined related modules

Tested how they work together
Checked data flow and communication
 Metrics:

State Transitions: Count transitions between different states

Measured processing time
Identified issues arising from component interactions

3.14.3 Full System Testing:

This evaluated the entire anti-collision system as a whole.

Procedure:

Tested the integrated system in a controlled environment

Simulated the system using Proteus software to real-world scenarios i.e., obstacle
detection, emergency braking
Assessed overall performance, accuracy, and reliability

Metrics:

Defect Density: Calculate defects per unit size.

Test Coverage: Measure the percentage of code exercised.
Pass Rate: Evaluate the success of test cases.
Response Time: Assess system reaction time.
Sensor Accuracy: Compare sensor measurements with ground truth3.

These metrics ensured the anti-collision system’s effectiveness and safety. By rigorously
testing and evaluating its performance
4 CHAPTER THREE
4.1 RESULTS AND DISCUSSIONS
This chapter discusses the results obtained for each objective
4.2 Objective one result
A table showing field statistical data on collision accidents for seven years

Number of NIGHT DAY Tunnel Others

Year
Accidents collision obstacles

2018 25 15 10 15 10

2019 30 25 5 13 17

2020 28 18 10 15 13

2021 29 15 14 15 14

2022 30 20 10 19 11

2023 40 26 14 17 13

2024 45 30 15 20 25

TOTAL 227 149 78 114 113

Data results on the condition of the trucks in the mechanical and in use that have been damaged
due to collisions in the tunnel.
Table showing the damaged truck parts
No TRUCK NUMBER Tick for damaged truck parts as a result of collision
NO. of damaged No. of damaged Side Grill guards
Head lamps mirror
1 UBM 142T 1 2 
2 UBM 302T x X X
3 UBM 090B 1 2 
4 UBM 781N X 1 
5 UBL 421K 1 2 X
6 UBL 703K X 1 
7 UBM 689X 1 2 
8 UBM 431S 1 X 
9 UBL 658A X 2 
10 UBL 117G 1 X 
11 UBM 631J X 2 
12 UBL 570H 2 1 X
13 UBM 006T X X 
14 UBL 034J 1 2 
15 UBM 583K X 1 
16 UBL 568Q X X 
17 UBN 312T 2 2 X
18 UBN 801B 1 2 
19 UBN 823S X 2 
20 UBN 057H 2 1 X
21 UBN 328H X 1 
22 UBL 863P 1 2 
23 UBL 974M X 2 
24 UBN 832F 1 1 
25 UBL 207B X X 
26 UAY 934M 1 1 X
27 UBL 100W X 2 
28 UAY 563U X 2 
29 UAY 873J 1 1 X
30 UAY 903G X 2 
31 UAY 410K X 2 
32 UAY 600J 1 2 
33 UAZ 422G X X X
34 UAZ861G 1 1 
35 UAZ 937Q 2 2 
36 UAZ 990S X 2 
37 UAZ738U 1 X X
38 UAZ 873R X X 
39 UAZ 095G 1 2 
40 UAZ 687M 

The results from the questionnaire of underground truck drivers show that 23 trucks (57.5%)
have damaged head lamps, 26 trucks (65%) have damaged side mirrors, and 34 trucks (85%)
have damaged grill guards, indicating that grill guards are the most common parts to be damaged
in collisions. Additionally, 6 trucks (15%) have no damaged head lamps, side mirrors, or grill
guards, and one truck (UBM 302T) has no damaged parts at all. Overall, the results suggest that
head lamps, side mirrors, and grill guards are the most susceptible to damage in collisions, with
grill guards being the most frequently damaged part
4.2.1 Data integration
Number of Accidents per year
50
45
40
35
30
Number of Accidents
25
20
15
10
5
0
1 2 3 4 5 6 7

Bar graph showing the relationship of night and day collision accidents

50
45
40
35
30
Number of Accidents
25
NIGHT
20 DAY
15
10
5
0
1 2 3 4 5 6 7

Integrated data from all the sources identifies key parameters contributing to collision accidents
as; Driver fatigue, inadequate safety protocols, Poor lighting system during night shifts.
 A line graph showing the trend of collisions between tunnels and
other objects
50
45
40
Number of accidents/yr

35 Number of Accidents
30 Tunnel collision
25 Others obstacles
20
15
10
5
0
1 2 3 4 5 6 7

The findings revealed key environmental factors contributing to collisions, including poor
visibility, narrow and congested roadways, weather conditions, road maintenance, lack of
signage and markings, noise and vibration (Nor Hidayah, 2015). These factors were also
triangulated to provide a comprehensive understanding of the environmental factors contributing
to collisions of trucks from underground mine drivers.
A review of 227 reported collision accidents on haul trucks for last 7 years was conducted and
identified 149 of the collision accidents occurred during night shifts, 50% of the accidents
involved haul trucks colliding with tunnel walls while negotiating a bend and other trucks, 20%
of the truck collisions were attributed to driver fatigue and 15% of the accidents occurred during
maintenance activities.
4.3 Objective two results
Table shows the comparison criteria weight of the criterias

C1 C2 C3 Criteria weight
C1 0.193548 0.264706 0.5 0.31941809
C2 0.322581 0.441176 0.2 0.321252372
C3 0.483871 0.294118 0.3 0.359329538
Bar graph showing the average weight of each criteria
0.4
0.35
0.3
0.25
0.2 Criteria weight
C1
0.15
0.1
0.05
0
1 2 3

4.3.1 Selection of the microcontroller

Criteria weight against criteria for microcontroller

0.5
0.45
0.4
0.35
0.3 Criteria weight
0.25
0.2
0.15
0.1
0.05
0
power consumption user friendly cost
Bar graph showing the weighted rating of the microcontroller

0.4

0.35

0.3

0.25

0.2
Series1
0.15

0.1

0.05

0
ARDUINO UNO Rasperry PI ARDUINO MEGA

Model graph relationships of obstacle distances, truck velocities and braking force
 A graph showing the relationship of varying model parameters

A graph showing the braking force and velocity for loaded and unloaded truck at a constant
braking distance
4.4 Objective three
REFERENCES
Abo-Hammour, Z.S., Alsmadi, O.M.K. and Al-Smadi, A.M. (2011) ‘Frequency-based model
order reduction via genetic algorithm approach’, 7th International Workshop on Systems, Signal
Processing and their Applications, WoSSPA 2011, (March 2015), pp. 91–94. Available at:
https://doi.org/10.1109/WOSSPA.2011.5931421.
Almeida, F. (2018) ‘Strategies to Perform a ixed Methods Study’, European Journal of
Education Studies, 5(1), pp. 137–151. Available at: https://doi.org/10.5281/zenodo.1406214.
Bareket, Z., Blower, D. F., MacAdam, C. (2000) ‘Blowout resistant tire study for commercial
highway vehicles’, 4(4).
Bhateria, R. and Jain, D. (2019) ‘Water quality assessment of lake water : a review’, Sustainable
Water Resources Management, 2(2), pp. 161–173. Available at: https://doi.org/10.1007/s40899-
015-0014-7.
Chakraborty, I., Tsiotras, P. and Diaz, R.S. (2013) ‘Time-optimal vehicle posture control to
mitigate unavoidable collisions using conventional control inputs’, Proceedings of the American
Control Conference, pp. 2165–2170. Available at: https://doi.org/10.1109/acc.2013.6580156.
Chehri, A. et al. (2020) ‘Realistic 5.9 GHz DSRC Vehicle-to-Vehicle Wireless Communication
Protocols for Cooperative Collision Warning in Underground Mining’, Smart Innovation,
Systems and Technologies, 185(May), pp. 133–141. Available at: https://doi.org/10.1007/978-
981-15-5270-0_12.
Dr. Vladimir, V.F. (1967) ‘Reducing Rollover Occupant Injuries: How and How Soon’,
Gastronomía ecuatoriana y turismo local., 1(69), pp. 5–24.
‘Faculty of Engineering - University of Porto SUSTAINABLE AND RESPONSIBLE MINING
BASED ON LUNDIN MINING CORPORATION ’ S SUBSIDIARY , SOMINCOR Master
Dissertation of’ (2022).
Giannetti, B.F. et al. (2020) ‘Insights on the United Nations Sustainable Development Goals
scope: Are they aligned with a “strong” sustainable development?’, Journal of Cleaner
Production, 252, p. 119574. Available at: https://doi.org/10.1016/j.jclepro.2019.119574.
Guenther, N. and Salow, H. (2015) ‘Collision avoidance and operator guidance - Innovating
mine vehicle safety’, Memmes 2015, pp. 1–10.
Highhouse, S. (2008) ‘Stubborn Reliance on Intuition and Subjectivity in Employee Selection’,
Industrial and Organizational Psychology, 1(3), pp. 333–342. Available at:
https://doi.org/10.1111/j.1754-9434.2008.00058.x.
Kecojevic, V. and Radomsky, M. (2004) ‘The causes and control of loader- and truck-related
fatalities in surface mining operations.’, Injury control and safety promotion, 11(4), pp. 239–251.
Available at: https://doi.org/10.1080/156609704/233/289779.
Middleton, N.J. (2017) ‘Desert dust hazards: A global review’, Aeolian Research, 24, pp. 53–63.
Available at: https://doi.org/10.1016/j.aeolia.2016.12.001.
Murshed, M. and Chowdhury, M.S. (2019) ‘An IoT Based Car Accident Prevention and
Detection System with Smart Brake Control’, (July), pp. 19–22. Available at:
https://www.researchgate.net/publication/333984519.
Na, D.E.C. and Hipertensiva, C. (no date) ‘No 主観的健康感を中心とした在宅高齢者における 健康関連指標に関する共分
散構造分析 Title’, (August 2023).
Naghshbandi, S.N., Varga, L. and Hu, Y. (2021) ‘Technologies for safe and resilient
earthmoving operations: A systematic literature review’, Automation in Construction, 125, pp.
1–31. Available at: https://doi.org/10.1016/j.autcon.2021.103632.
Nor Hidayah, M.A. (2015) ‘Impact of Industrial Effluents on Water Quality At the Tiram’,
(June).
Patrucco, M. et al. (2021) ‘Anti-collision systems in tunneling to improve effectiveness and
safety in a system-quality approach: A review of the state of the art’, Infrastructures, 6(3).
Available at: https://doi.org/10.3390/infrastructures6030042.
Rahimdel, M.J. et al. (2014) ‘Maintenance plan for a fleet of rotary drill rigs’, Archives of
Mining Sciences, 59(2), pp. 441–453. Available at: https://doi.org/10.2478/amsc-2014-0031.
Rakhonde, M.A., Khoje, S.A. and Komati, R.D. (2018) ‘Vehicle Collision Detection and
Avoidance with Pollution Monitoring System Using IoT’, Proceedings - 2018 IEEE Global
Conference on Wireless Computing and Networking, GCWCN 2018, (November 2018), pp. 75–
79. Available at: https://doi.org/10.1109/GCWCN.2018.8668622.
Reiman, A. (2020) ‘Human factors and maintenance in delivery transportation: drivers’ work
outside the cab in focus’, Journal of Quality in Maintenance Engineering, 27(3), pp. 465–482.
Available at: https://doi.org/10.1108/JQME-05-2020-0035.
Ripka, P. and Tipek, A. (2010) Modern Sensors Handbook, Modern Sensors Handbook.
Available at: https://doi.org/10.1002/9780470612231.
Rogers, W.P. et al. (2019) ‘Automation in the Mining Industry: Review of Technology, Systems,
Human Factors, and Political Risk’, Mining, Metallurgy and Exploration, 36(4), pp. 607–631.
Available at: https://doi.org/10.1007/s42461-019-0094-2.
Sachdeva, S. (2019) ‘SENSORS BASED EARLY WARNING SYSTEM’, 6(4), pp. 1411–1414.
Siecker, J., Kusakana, K. and Numbi, B.P. (2017) ‘A review of solar photovoltaic systems
cooling technologies’, Renewable and Sustainable Energy Reviews, 79(July 2016), pp. 192–203.
Available at: https://doi.org/10.1016/j.rser.2017.05.053.
Sindi, S. and Woodman, R. (2021) ‘Implementing commercial autonomous road haulage in
freight operations: An industry perspective’, Transportation Research Part A: Policy and
Practice, 152, pp. 235–253. Available at: https://doi.org/10.1016/j.tra.2021.08.003.
Wang, X. et al. (2016) ‘Development of a Kinematic-Based Forward Collision Warning
Algorithm Using an Advanced Driving Simulator’, IEEE Transactions on Intelligent
Transportation Systems, 17(9), pp. 2583–2591. Available at:
https://doi.org/10.1109/TITS.2016.2528508.
Zhang, M. (2014) ‘Analysis of Haul Truck-Related Fatalities and Injuries in Surface Coal
Mining in West Virginia’. Available at:
https://researchrepository.wvu.edu/etd/7031/%0Ahttps://researchrepository.wvu.edu/cgi/
viewcontent.cgi?article=8072&context=etd.
Zhang, M., Kecojevic, V. and Komljenovic, D. (2014) ‘Investigation of haul truck-related fatal
accidents in surface mining using fault tree analysis’, Safety Science, 65, pp. 106–117. Available
at: https://doi.org/10.1016/j.ssci.2014.01.005.
Zhang, Y., Antonsson, E.K. and Grote, K. (2006) ‘A new threat assessment measure for collision
avoidance systems’, IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC,
(October 2006), pp. 968–975. Available at: https://doi.org/10.1109/itsc.2006.1706870.
Zhuang, Y.-Y. et al. (2023) ‘The application of intelligent sensors in medical research: a review’,
Biomedical Engineering Communications, 2(3), p. 17. Available at:
https://doi.org/10.53388/bmec2023017.
Appendix
Arduino code for determining servo motor deflection in relation to required brake pad
force
#include <Servo.h>
// Define servo motor pins
const int clutchServoPin = 9;
const int brakeServoPin = 10;
// Define servo motor objects
Servo clutchServo;
Servo brakeServo;
// Define maximum braking force and motor deflection
const int F_max = 360; // N
const int x_max = 180; // degrees (both clutch and brake)
void setup() {
// Initialize servo motors
clutchServo.attach(clutchServoPin);
brakeServo.attach(brakeServoPin);}

void loop() {
// Activate clutch servo to disengage system
clutchServo.write(x_max); // full deflection (180 degrees)
delay(1000); // wait 1 second
// Calculate brake pull force based on relationship
int F = (F_max / x_max) * x_max; // full deflection (180 degrees)
// Activate brake servo to apply pull force
brakeServo.write(x_max); // full deflection (180 degrees)
delay(1000); // wait 1 second
}