DRIVER ALERT MECHANISM AND SAFE TRACTION OF VEHICLE ON THE ROAD

Raspberry and Arduino uno

Ameera Ali Jomin Joy

Department of Electrical and Electronics Engineering Department of Electrical and
Electronics Engineering Amal Jyothi College of Engineering Amal
Jyothi College of Engineering
Kerala Technical University Kerala Technical University

Joyel N Thomas Tivin John Itty

Department of Electrical and Electronics Engineering Department of Electrical and
Electronics Engineering Amal Jyothi College of Engineering Amal
Jyothi College of Engineering
Kerala Technical University Kerala Technical University

INTRODUCTION
“Drowsiness is a silent killer on the roads, and it’s important to take proactive
measures to prevent it”
Loss of driver alertness is almost always preceded by psycho-physiological changes; these changes
are the reason that it is possible to detect the onset of drowsiness associated with loss of alertness
in driving. Driver lethargy and drowsiness are substantial causes of accidents on roads worldwide.
According to the National Highway Traffic Safety Administration (NHTSA), approximately 100,000
crashes each year are caused by driver fatigue, resulting in around 1,550 deaths and 71,000 injuries.
Several driver alert mechanisms have been developed to reduce the number of accidents caused by
driver tiredness. The research paper focuses on the development of driver alert mechanisms using
Arduino and an eye blink sensor. The proposed system detects the driver’s level of drowsiness by
monitoring their eye-blink patterns and provides an alert to the driver, initially decelerating the
vehicle and finally stopping. This research paper aims to evaluate the effectiveness of the proposed
system in reducing the number of accidents caused by driver fatigue.
As we know that nowadays vehicle is designed on the methodology of CRDI [Common Rail Direct
Injection] and MPFI [Multi Point Fuel Injection], fuel injection systems if the driver falls asleep the
foot from the pedal is released because of the related mechanism of the body, but in the most
modern technology fuel have been given to engine by fuel pump this pumping continues even we
take the foot from the accelerator pedal due to moment of inertia of the vehicle.
This system can be applicable in passenger vehicles and carriers to prevent homicide. Suppose a
truck transporting flammable things, due to driver drowsiness, can cause a road accident and an
explosion that can kill a few people. It is vital to design and develop the most modern advanced
driver alert mechanism to overcome all these hazards.

LITERATURE REVIEW
Possible techniques for defecting drowsiness are generally divided into two main categories. The
first category is related to physiological studies. It includes eye-lid closure, facial recognition, heart
rate monitoring, electrooculogram (EOG) and Electroencephalogram (EEG). Although the EOG
provides good results, it is not suitable for a real driving environment because of the attached
electrodes on the driver’s face. The second category is focused on the driving characteristics and
unusual behaviour of the vehicle such as the use of reaction time of the driver, lane departure
sensing, accelerator Or brake pattern, lateral acceleration, lateral displacement and steering wheel
movements. Among other methods, the eye-blink duration is the most reliable parameter for the
detection of the drowsiness level. Drowsy driver detection system[1] using eye blink pattern by
Taner Danisman, Ian Marius Bilasco, Chabane Djeraba, and Nacim Ihaddadene proposed a method
to detect drowsy by analysing visual changes in eye location using the horizontal symmetry feature
of the eye. In this study, they present a low-cost and fully automatic solution for handling the
drowsy driver detection problem. Their system uses a standard webcam and detects the pattern of
long-duration eyelid closures. In this system when a driver falls asleep, the driver gets alerted by
voice messages. But there is a chance that the driver may not hear the voice message. In that case,
giving an alert to the driver by voice message is not applicable. Neeta Parmar’s drowsy detection
system focused on the localization of the eyes, which involves looking at the entire image of the
face and determining the position of the eyes by a self-developed image-processing algorithm.
After locating the eyes, the system determines whether the eyes are opened or closed, and detect
fatigue. This system also deals with drowsy detection only. But our aim is to stop the vehicle
without any danger. IOT-based Real-time Drowsy Driving Detection System [3] for the Prevention of
Road Accidents by Yousuf Hossain and Fabian Parousia George proposed a method to detect
drowsily. This system used the eye closure ratio as an input parameter to detect the drowsiness of
the driver. If the eye closure ratio decrease from the standard ratio, the driver is alerted with the
help of a buzzer. A Pi camera is used in this system to capture the images of the driver’s eye and the
entire system is incorporated using Raspberry Pi. However, there is a drawback in this system,
which is its incapability to serve its purpose at night. Seok-Woo Jang’s and Byeongtae Ahn’s
Implementation of Detection System for Drowsy Driving Prevention Using Image Recognition and
IOT: In this study, machine learning was applied to predict drowsiness and improve drowsiness
prediction using facial recognition technology and eye-blink recognition technology. Additionally, a
CO2 sensor chip was used to detect additional drowsiness. Speech recognition technology is also
used to apply Speech to Text, it allows the driver to request their desired music or make a call to
avoid drowsiness while driving. In addition, they used Python and C languages for its development
and it used a Raspberry Pi 3, infrared camera, speaker, microphone, carbon dioxide sensor, Galaxy
S4, and the automobile model Sonata. However, there is no external warning system or parking
system included.

PROPOSED METHODOLOGY
The proposed design here is designed to reduce the dreadful accidents that are happening
throughout the world. Nowadays carelessness and fatigue cause accidents. With their busy
schedules, people’s lack of sleep has become a major issue for accidents as well as a bad impression
on health. Lack of sleep can cause the brain to transfer false neuron signals.To overcome these by
analysing different behaviours of eyelids with the help of an eye blink sensor(raspberry pi cam). This
sensor continuously monitors and studies the movements of the eyes, providing signals to the
microprocessor system(raspberry pi). These signal gets examined to understand the actual state of
the driver.After a certain period of time (a few seconds<3sec), if the driver’s eyelid is closed more
than the normal time, the sensor gets activated and stops the vehicle through slow retardation and
finally stops.But the real problem here is that the lack of proper warnings to the nearby drivers may
cause mutliaccidents. So it requires proper parking of the vehicle on the roadside. This can be
achieved by combining two different types of sensors: a pair of ultrasonic sensors to find objects
and ZX distance and a gesture sensor, a touchless sensor capable of detecting simple gestures. It
can recognize the distance of an object 30cm away on the Z axis, and 15cm on the X axis. This
gesture sensor helps to identify the presence of humans on roadsides.Mainly focus on the traction
of the vehicle in safe parking. This perpendicular motion of the vehicle is achieved by mecanum
wheels.By controlling each motor we can achieve different motions of the vehicle such as
perpendicular, circular etc.

Figure1. Pictorial representation of movements of mecanum wheels

RESUTS AND DISCUSSION

Driver lethargy is a meaningful question on roads in general.It is individual of important causes of
course accidents and fatalities.Drowsiness influence the jockeys talent to uphold consideration
respond to surprising occurrence form speedy resolutions.Drowsy forceful is usually induced by
sleep needs,long active hours, cure and healing environments to a degree of sleep interruption of
activity.
Research has proved that sleepy forceful is more coarse all the while darkness first glance hours and
following in position or time middle of the day.Drivers the one are sleep-impoverished or have a past
of sleep disorders are at taller risk of being sleepy forceful.Additionally operator the one work
midnight shifts or have long occupied hours are again more inclined happening operator lethargy.To
prevent lethargy,masters approve of clutching adequate sleep before forceful,attractive breaks every
two hours all the while long drives preventing in intoxication and sedatives before forceful and
pursuing healing considerations if you have a sleep disorder.Furthermore,electronics to a degree
sensors and camera may be equipped in buses to discover sleepy forceful and alert chauffeurs to take
breaks. Conclusion trainer lethargy is a weighty issue that’s demands considers from two together
chauffeurs and policymakers.Increased knowledge campaigns, instruction and electronics can help
counter sleepy force and defeat the number of accidents precipitated by it.From the above statements
and findings in the literature review we come to know that there were preexisting mechanism to give
warnings and alertness To drivers so as to stop the vehicle but it is on the way(road itself),so it is
necessary to develop a more efficient mechanism that can automatically park the park the vehicle to
the side of the road.After a lot of observation and studies based on research papers and reference
source, we finally arrived at the use of raspberry pi and raspberry pi cam as eye blink sensors (of
94% accuracy) and the meccanum wheel for the traction of vehicle with help of some sensors
modules to park it safely.The raspberry pi cam senses the eyelid blinking and sends signals to the
raspberry which can pass over to the ardunio uno through the transmitters and receiver ports of both
micro-controllers.The ardunio gives the signal to the gear motors,thus enabling the futher activations
of sensors to take place.
CIRCUIT AND WORKING MECHANISM
This project is mainly for drivers especially who are driving a long distance or for those drivers who
are taking a night drive. Here we use a motor as our engine to demonstrate the working. This is
mainly worked on Raspberry, Arduino uno and the eye blink sensor(raspberry pi cam).The above-
listed components together function to achieve the developed design. On the dashboard in front of
the driver the Pi cam is fixed so that it can get a better facial monitorization of the behavioural
characteristics. On sensing the movements it continuously sends signals to the raspberry pi. On
studying the signals appropriate signal is passed between the transceiver ports[TX-RX] of the
raspberry pi and the Arduino Uno.The requested signal from the Raspberry to Arduino begins to
decelerate the motor that is the mecanum wheels and finally to rest. As soon the signal is
generated from the Arduino the hazard indicators start to glow so that it can warn the nearby
drivers. Since the vehicle is on the road we need to park it safely. For that zx sensors and gesture
sensors, ultrasonic sensors, and line follower sensors get activated. They collect the data and
compute the best coordinate position to park the vehicle perpendicularly from the road.
Figure.2

Figure.3

Here we designed a line follower that can detect the white line available on the road. The
perpendicular motion or the circular motion or the diagonal motions of the vehicle can be
get by applying different working combinations of the motor .
The IoT-based application is being developed through the integration of some IoT modules
like sensors, a Pi camera, and smart code for detecting the drowsiness of the driver, even a
Bluetooth module is optioned to connect it with the mobile phone. So the above modules
are properly integrated with the Raspberry Pi controller module and Arduino uno that
intelligently controls and smartlwarnsaa drowsy driver and take control over the driver
considering the safety and value of life.

Figure.4
BLOCK DIAGRAM

COMPONENTS DESCRIPTION
The Arduino UNO:

The Arduino UNO is a programmable micro-

controller board that can be incorporated
into a variety of electronic projects. It is
open-source, inexpensive, flexible, and
simple to use. The board can control relays,
LEDs, servos, and motors as outputs and can
be connected to other Arduino boards,
Arduino shields, and Raspberry Pi boards.

Figure.5

L298N Motor Operator:

Figure.6

Motors with voltages ranging from 5 to 35V DC can be used with the H-bridge module L298N. The
integrated circuit L298N is a stand-alone unit. It is a dual global driver with a high voltage and a high
current that can drive inductive loads like relays, solenoid coils, DC motors, and stepper motors at
standard TTL logic levels.

ZX Distance and Gesture sensor:-

Figure.7

The contactless gesture and

proximity sensor ZX can track simple gestures. The "Z" axis, which is the distance an object is from
the sensor up to 12 inches, is measured by the GestureSense technology used by the sensor. In
order for the ZX sensor to function, two LEDs on either side of the object that is above it bounce
infrared (IR) light beams. A micro-controller on the back of the sensor interprets the data as the
reflected light returns to the receiver in the middle of the sensor.
IR White line follower:-

The line adherent IR sensor can consequently follow the dark or

white line on a superficial level without outside control. A photo-
transistor and an infrared LED are positioned side by side. The Drove
goes about as a transmitter and the photo-transistor goes about as a
recipient.

Figure.8

Raspberry Pi 3 Model B Board:-

Figure.9

This board is Built on the latest Broadcom 2837 ARMv8 64bit processor. The new generation
Raspberry Pi 3 Model B is faster and more powerful than its predecessor models. The installed
memory on this board is 1 GB. This board is a quad-core board with a processor speed of 1.2GHz.Its
Dimensions are 12.19 x 7.59 x 3.4 cm; 45 Grams

Raspberry Pi camera:
 Figure.10

The Raspberry Pi Camera Board is a custom module that can be added to

the hardware of the Raspberry Pi. utilizes a specialized CSI interface to connect to the Raspberry Pi
hardware. In still image capture mode, the sensor has a native resolution of 5 megapixels. Video
mode can record at 30 frames per second up to 1080p resolution.
Double engines: (dual pair of the motor)

Figure.11

Four-wheel drive with two motors is achieved by mounting two engines—one front and one rear—
on a single chassis. The front motor sends capacity to the front wheels, while the back motor sends
capacity to the back tires, conveying a phenomenal speed increase.
Mecanum wheels :-

Figure.12

The mecanum wheel is an omnidirectional wheel plan that permits the vehicle on the set to move
toward any path. It is a resolute wheel, with a progression of elastic external rollers skewed around
the whole circuit of the edge. The axis of rotation of these rollers is typically 45 degrees from the
shaft line to the wheel plane
Figure.13

REFERENCES
 G. Pan, L. Sun, Z. Wu and S. Lao, “Eyeblink-based Anti-spoofing in Face Recognition from
a Generic Web camera,” The 11th IEEE International Conference on Computer Vision
(ICCV'07), Rio de Janeiro, Brazil, October 2007.
 Seok-Woo Jang and Byeongtae Ahn-Implementation of a Detection System for Drowsy
Driving Prevention Using Image Recognition and IoT in Department of Software, Anyang
University, Liberal and Arts College, Anyang 14028, Korea;
Published: 10 April 2020
 Taner Danisman, Ian Marius Bilasco, Chabane Djeraba, Nacim Ihaddadene- Drowsy Driver
Detection System Using Eye Blink patterns -LIFL UMR CNRS 8022, Université Lille 1 &
Telecom Lille 1, Villeneuve d'Ascq 59655, Lille, France
 Md. Yousuf Hossain and Fabien Parsia George-
IoT-based Real-time Drowsy Driving Detection System for the Prevention of Road
Accidents, Department of Computer Science and Engineering
BRAC University
Dhaka, Bangladesh
 Ronald R. Knipling and Walter W. Wierwille-Vehicle-Based Drowsy Driver Detection:
Current Status and Future Prospects, Office of Crash Avoidance Research Vehicle Analysis
and Simulation Laboratory
National Highway Traffic Safety Administration Virginia Polytechnic Institute and State
University
 Kyung-Lyong Han, Oh Kyu Choi, In Lee, Inwook
Hwang, Jin S. Lee, and Seungmoon Choi, “Design
and Control of Omni-Directional Mobile Robot for
Mobile Haptic Interface”, International Conference
on Control, Automation and Systems 2008, pp. 1290–1295, 2008.
 Y.P. Leow and K.H. Low and W.K. Loh, “Kinematic
modelling and analysis of mobile robots with omni-directional wheels”, Control, Automation,
Robotics
and Vision, 2002. ICARCV 2002. 7th International
Conference on, Vol. 2, pp. 820–825, 2002.