Proceedings of the 2nd International Conference on Trends in Electronics and Informatics (ICOEI 2018)

IEEE Conference Record: # 42666; IEEE Xplore ISBN:978-1-5386-3570-4

Design and Implementation of Intelligent Headlight

system for improvised safety in vehicles
A.Guru Dineshkumar Abhay kumar Singh Karthikeyan. B
School of Electronics School of Electronics School of Electronics
Engineering Engineering Engineering
VIT Vellore, India VIT Vellore, India VIT Vellore, India
[email protected] [email protected] [email protected]
om

Abstract—Traditional headlamps fail to provide the optimal

illumination on road during nighttime. This paper is focused on
design and working of Intelligent Headlight System (IHS) using
Raspberry Pi 3. Here steering of the vehicle is considered as a
potentiometer or an accelerometer, the servo motors along with
LEDs are rotated accordingly to the steering angle produced
from the data provided by potentiometer or accelerometer so as
allowing light to spread over the entire road. LDRs, Pi Camera
used to detect light from other vehicles and utilized to reduce the
effect of glare by controlling the intensity of light.

Keywords—Potentiometer, Accelerometer, Servomotor, LED,

LDR, Raspberry Pi 3, Pi Camera, Python, Open CV, C++.

I. INTRODUCTION Figure 1: With and without Adaptive headlight effect

Ordinary headlamps have a problem providing better
illumination on road during nighttime i.e. because they are
stern and while at turnings, the entire road is not visible for the
driver hence it is dangerous and causes many road accidents.
Road accidents are ranked as the ninth leading cause of deaths
i.e., around 2.2% globally.
According to the recent studies, Literature [1] 82% of the
road accidents occurred during the night due to poor visibility.
Around 50% of those road accidents occur at corners as the
vehicle is not able to adapt to sudden change in the direction, Figure 2(a) Figure 2(b)
creating a condition of “blind spot” which hampers the driver’s
field of vision while turning the corner. Literature [2] shows Fig. [2(a)] depicts the reduced glare after implementing the
that around 30% of drivers are facing a problem with the glare, IHS. Fig. [2(b)] has glare effect, which decreases the visibility
which causes a large number of havocs on road each year, Also of driver and causes accidents.
this number is larger in drivers aged above 45 as they have
poor sensitivity to glare, due to low contrast when light enters As of now, there are many proposed models of Adaptive
the eye. Lighting system [8]. Keeping those in mind a combined
intelligent control system [9] is designed which caters to driver
The headlamps that are presently used are not providing
assistance. In [3], they designed and built a model of turntable
sufficient illumination on curved roads. Due to improper
headlamps by adapting traditional static headlamps. Earlier, [4]
visibility at turnings at nighttime, many road accidents are
they employ the method of using steering wheel angle and
occurring. To avoid road accidents there are many proposed
speed to determine the angle rotated by the headlamps.
models of Adaptive Front lighting System (AFS).
However, there is lag produced by the CAN in the vehicles [4].
Fig. [1] Shows the comparison of a normal car and a car with
In this work, there were derived mathematical equations for
Intelligent Headlight System (IHS). It is clearly visible that on
finding turning angle in vertical and horizontal directions and
a cornered road the sight of the driver is not accessible to the
the design of AFS made accordingly [5]. Body Control Module
right side of the road making it a complete “blind zone” [12].
(BCM) calculates turning radius. From the data of turning
radius, swivel angle is calculated. From the data available,
swivel angle of headlamps is controlled [6].

978-1-5386-3570-4/18/$31.00 ©2018 IEEE 667

Authorized licensed use limited to: ST. JOSEPH ENGINEERING COLLEGE MANGALORE. Downloaded on November 14,2024 at 10:54:02 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the 2nd International Conference on Trends in Electronics and Informatics (ICOEI 2018)
IEEE Conference Record: # 42666; IEEE Xplore ISBN:978-1-5386-3570-4

(6) ADC (Analog-to-Digital converter MCP-3008):

Raspberry Pi does not have any ADC inbuilt hence; A/D
converter is required for doing so. MCP-3008 is 8-channel
10-bit analog to digital converter having an SPI (Serial
Peripheral Interface).
(7) LED Headlamps:
LEDs offer a brighter, higher quality lighting. The amount
of power consumption, when compared to others, is quite
small. They are cheaper than the older halogen headlamps;
also, they are smaller which helps in controlling the
individual LEDs providing greater feasibility and many
more functions.
(8) LDR (Light Dependent Resistor):
LDR works on how much amount of light is falling on the
surface. When the level of light is more the resistance
Figure 3: Overview of system design generated by LDR decreases and vice-versa. The intensity
of light falling onto the car is determined by using LDR.
Additionally, to reduce the glare caused by the other
II. DESIGN vehicles, reduce the intensity of LEDs.

Requirements (9) Operating System:

(1) Raspberry Pi 3: Raspberry Pi requires an operating system to interact with
the peripherals. Raspbian stretch is a Debian based OS for
Raspberry Pi 3 is the single board computer, which acts as Raspberry Pi.
the main processor by continuously tracking the steering
wheel angle via potentiometer and accelerometer.
Movement of the servomotors controlled by manipulated III. METHODOLOGY AND IMPLEMENTATION
angle, which eventually controls the direction of the
headlights. In order to control the glare, it uses Pi Camera The whole architecture of IHS divided into 2 major parts:
and LDR for controlling the intensity of the LED
Headlamps. (1) Adaptive control of light angle

(2) Pi Camera: (2) Intensity control of headlamps

Pi camera is the most suitable camera for image processing The LED lights attached to the Servomotors via an
via Raspberry Pi. Pi Camera used in monitoring the road for attachment, which helps in the swivel movement. Now first
any vehicle and tracking the distance from it. In addition, the Rotatory Potentiometer gives out the values according
this assists to find the maximum intensity of light in that to the rotation whether left or right. First, that reading sent
area. to MCP-3008 ADC to convert it to the digital values for
processing by Raspberry Pi. The MPU6050 module
(3) Potentiometer:
provides the YPR (Yaw, Pitch, and Roll) which when
A 10 k rotatory potentiometer acts as the steering wheel in combined with the potentiometer data give the exact
the implementation. Even in ECU (Electronic control unit) rotation of car on corners. This change in steering
[10]
, a rotatory potentiometer is attached at the end of perceived by Raspberry Pi and that gives the PWM signal
steering column. to the corresponding servomotors to turn. While turning
(4) Servomotors (SG-90): one servomotor will have to move a larger degree when
compared to other, that is decided by the direction, as we
Servomotors used for movement involving higher precision know that from [5], the degree of rotation in the horizontal
and smaller divisions. These motors turn to a certain angle direction for LEDs is +150 to -150, while in the vertical
depending on the PWM (Pulse Width Modulation)
duration. Mostly they turn total angle of 1800 in total. direction it’s 0 to 70.
Next module consists of LDR, which gives the output in
analog, here we cannot use the analog values directly and
(5) Accelerometer and Gyroscope (MPU6050): hence ADC is required for getting the resistance value, as
This module helps in finding the rotation angle w.r.t a static one channel used for rotatory potentiometer, 2 more
car, and act as the input to the Raspberry pi for YPR (Yaw, channels are for LDRs. These LDRs are kept facing on
Pitch, and Roll) to use this data further for swiveling of sides of the car, which continuously keep track of light
motors along with potentiometer.

978-1-5386-3570-4/18/$31.00 ©2018 IEEE 668

Authorized licensed use limited to: ST. JOSEPH ENGINEERING COLLEGE MANGALORE. Downloaded on November 14,2024 at 10:54:02 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the 2nd International Conference on Trends in Electronics and Informatics (ICOEI 2018)
IEEE Conference Record: # 42666; IEEE Xplore ISBN:978-1-5386-3570-4

intensity. There are different levels of brightness in LEDs, location[i] gives the location of the contour box, which
which can be set using the PWM time cycle. keeps on searching for the perfect match of a Car in the
frame processing.
As an upgrade, a camera [7] used to find the vehicle, track it
and continuously find a distance value, which uses Haar
Cascaded classifier for recognizing the car. The coding
done is in C++ language. After getting the result as the
distance in meters, we go on for controlling the servo motor
and LED intensity. There is distance threshold, which
maintains and looks for new objects in that range only.
Several image-processing techniques used to arrive at the
final frame result.
Figure 5: Region of interest in video
IV. RESULTS

Figure 6: Binary Canny Edge Detection

Figure 4 : Hardware requirements for the project

Fig. [4] Shows the actual hardware of the project. As shown

in the image the servomotors attached to the LEDs whose
intensity controlled by the LDR and the Pi Camera installed
on the top of the box.
The next Fig. [5] Shows the various filtering done to get to
the final tracking and detection of the car. The first image
shows the Region of interest that checks persistently for
vehicles. Next is the canny edge detection, that gives a
binary image, having edges about the road features. Figure 7: Final video frame tracking vehicle and distance
Next is the Output frame of the processed video by the
Raspberry Pi. This frame video calculates the distance Finally using this distance from the car in the frames PWM
using a Haar cascaded classifier that detects the features of Signals sent to the servomotors control their rotations and
the vehicle and compares them to the hundreds of models in use the LDR for the brightness level in case there is a
an XML file. vehicle approaching nearby.
After comparing, them the final distance found out by using Fig. [8] Shows the turned angle of servomotor when the
a default value of width of the cars using a simple equation. steering wheel angle sensor (potentiometer or
Distance= (0.0397*2)/((locations[i].width)*0.00007) Where accelerometer) rotated to one side. The servomotor seemed
in the factor (0.0397*2) gives an average width of a car and

978-1-5386-3570-4/18/$31.00 ©2018 IEEE 669

Authorized licensed use limited to: ST. JOSEPH ENGINEERING COLLEGE MANGALORE. Downloaded on November 14,2024 at 10:54:02 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the 2nd International Conference on Trends in Electronics and Informatics (ICOEI 2018)
IEEE Conference Record: # 42666; IEEE Xplore ISBN:978-1-5386-3570-4

to be turning in the same way to provide illumination in the V. CONCLUSION AND FUTURE WORK
“blind spot” region. In this paper, we mainly focus on assisting the driver during the
night by providing proper illumination on the road at turnings
especially with the help of servomotors along with LEDs and
reducing the glare effect from the oncoming vehicles by using
LDR or Pi Camera for detection. The intensity of headlamps
controlled based on the distance of the oncoming vehicles.
There are certain modes provided based on the speed of the
vehicle to control the shape and intensity of light from
headlamps. We are working towards recognizing the road
signs, traffic signals, pedestrians to ensure safety and this work
will make improvement in ADAS (Advanced Driver
Assistance System). This kind of system will be helpful in
Autonomous cars.

REFERENCES
[1] Design & Implementation of Adaptive Front Light System of Vehicle
Using FPGA Based LIN Controller)
[2] J. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol.
2. Oxford: Clarendon, 1892, pp.68–73.
[3] Meftah Hrairi and Anwar B. Abu Bakar, “ Development of an Adaptive
Headlamp Systems”, International Conference On Computer And
 Communication Engineering (ICCCE 2010), 11-13 May 2010, Kuala
Lumpur.
Figure 8: Hardware System Design [4] Steering Wheel-based Adaptive Headlight Controller with Symmetric
Angle Sensor Compensator for Functional Safety Requirement. Jiae
Youn, Meng Di Yin, Jeonghun Cho, and Daejin Park.
[5] Yali Guo, Qinmu Wu And Honglei Wang, “ Design and Implementation
of Intelligent Headlamps Control System based on CAN Bus”, 2012
International Conference on Systems and Informatics (ICSAI2012).
[6] Hyeongyun Moon, Jeonghun Cho, and Daejin Park “Effective Validation
Framework Using Wheel-Microcontroller Co-Emulation of Safe
Adaptive Front Light Controller”.
[7] Snehal G. Magar. “Adaptive Front Light Systems of Vehicle for Road
Safety”.
[8] Xu pingping, Song jianguo, Shen Guangdi. “The design on Adaptive
Front Lighting System (AFS) Based on Brushless DC Motor”.
[9] Guo Dong, Gao Song, Wang Hongpei, Wang Jing. “Study on Adaptive
Front-lighting System of Automobile Based on Microcontroller”.
[10] Jiae Youn , Men Di Yin , Jeonghun Cho , and Daejin Park “Embedded
ECU Emulation Platform for Safe Adaptive Front-Light System Control
Algorithm Development”.
[11] Jun Ho Lee, Jina Byeon, Dong Jin Go and Jong Ryul Park “Automotive
Adaptive Front Lighting Requiring Only On/Off Modulation of Multi-
array LEDs”.
[12] Li Lifu, Yang Mingjun and Zhang Jinyong, “The Bending Mode Control
Method Of Afs System Based On Preview Control”.

978-1-5386-3570-4/18/$31.00 ©2018 IEEE 670

Authorized licensed use limited to: ST. JOSEPH ENGINEERING COLLEGE MANGALORE. Downloaded on November 14,2024 at 10:54:02 UTC from IEEE Xplore. Restrictions apply.