2020 2nd International Conference on Industrial Electrical and Electronics (ICIEE)

Design of Automatic Headlight System Based on

Road Contour and Beam from Other Headlights
Fadil Muhammad Dzulfiqar Dwi Yanto Wahyuni Martiningsih
Department of Electrical Engineering Department of Electrical Engineering Department of Electrical Engineering
Universitas Sultan Ageng Tirtayasa Universitas Sultan Ageng Tirtayasa Universitas Sultan Ageng Tirtayasa
Cilegon, Indonesia Cilegon, Indonesia Cilegon, Indonesia
fadil.muhammad@untirta.ac.id dzuditokun@gmail.com y martingsih@untirta.ac.id

Vembi Noverli Romi Wiryadinata

Faculty of Engineering, Transport Systems and Logistics Department of Electrical Engineering
Universität Duisburg-Essen Universitas Sultan Ageng Tirtayasa
Duisburg, Germany Cilegon, Indonesia
vembi.noverli@stud.uni-due.de wiryadinata@untirta.ac.id

Abstract—Accidents often occur at night due to a lack of light. technology we call the smart car, which is a car that combines
With the low-light conditions, a lot of drivers who use high beam electrical, electronic, and communication technology [4] - [5].
headlight forgot to switch to low beam headlight. That action can Smart car development aims to improve the safety and comfort
cause temporary blindness to the driver in front of him because of
the glare. Automatic Headlight can change the mode of lights and of vehicle users.
reflector lights when passing move uphill or downhill roads using
the accelerometer sensor MPU6050. The automatic headlight can II. R ELATED W ORKS
also change the mode of light based on the light in front of him Previous research has designed a car that can detect the
using the BH1750 lux meter sensor so that the path illuminated condition of the road surface [6] - [7]. The research uses
by the headlight to be quite and do not endanger other drivers.
Laboratory testing has a success rate of 97%, with 260 trials in the Inertial Measurement Unit (IMU), which utilizes an ac-
13 conditions. And on real testing, the automatic headlight can celerometer and gyroscope to measure the road surface. Other
switch modes automatically when going uphill and downhill. The studies have discussed methods for controlling headlight by
headlight can change modes automatically when passing through adjusting the speed and tilt of the vehicle [8]. Other research
roads with adequate lighting or when passing other vehicles. also discusses how to detect other vehicles at night using
Index Terms—Automatic Headlight, MPU6050 Accelerometer,
Luxmeter BH1750
cameras and image processing methods [9]. Similar research
also discusses how to identify other vehicles using a camera
with a curb line as a reference [10]. The disadvantages of
I. I NTRODUCTION
these studies are not suitable for use in the way that there is
Car ownership in the last two decades has spread rapidly no curb line. Most of the previous studies used the camera as
at every level of society, so safety in driving is a significant the main component to detect the presence of other vehicles.
concern. Some of the factors that cause traffic accidents are The use of the camera as the main component is less efficient
humans, vehicles, roads, and the environment. Vehicle factors when viewed from a financial perspective. Besides, the use of
contributed 2.76% in traffic accidents in Indonesia [1]. Most image processing techniques is also considered less efficient
accidents occur at night, especially if there is no street lighting in terms of computing, if only to replace the headlight.
on the road [2]. The misuse of high beam lights causes some This study uses the BH1750 module as the main component
accidents that occur at night [3]. When faced with the other to detect other vehicles facing in front of it. This study also
vehicles, drivers forbidden to turn on the high beam lights used the MPU6050 module to assess road contours. That way,
because when the vehicle meets each other, another driver the Headlight not only changes the automatic mode based
who receives the high beam light dazzled and can experience on the presence of other vehicles but can also change when
momentary blindness that can cause accidents. Many of the passing the uphill or downhill road. So we need a system
drivers forget to change the vehicle’s beam headlight mode that can reduce the direction of the headlight beam when
when passing on dark uphill or downhill roads. The habit of passing through the uphill road, can increase the course of the
not changing the headlight mode is dangerous for other drivers headlight beam when the road is down, and automatic light
and the driver himself. For that, we need a system that can modes that can be automatically changed when facing other
change the car’s headlight mode automatically. The entry of vehicles. When passing the road uphill, the vehicle load point
electronic devices into motorized vehicle systems has changed will be centered on the rear. Conversely, the car’s load point
the direction of development and innovation in vehicles. The on the downhill road will be focused on the front so that the

978-1-7281-9702-9/20/$31.00 ©2020 IEEE 112

Authorized licensed use limited to: Auckland University of Technology. Downloaded on December 20,2020 at 15:55:27 UTC from IEEE Xplore. Restrictions apply.
 2020 2nd International Conference on Industrial Electrical and Electronics (ICIEE)

back of the vehicle rises and causes the Headlight beam to be

closer. This condition causes the front of the car to lift. So
that the headlight beam goes up and does not illuminate the
road properly can even cause other drivers to glare.
III. P ROPOSED M ETHOD

Fig. 1. Block diagram of the system.

This research design and create automatic headlight based

contours of the road and other Headlight lights. The block
diagram in Fig. 1 shows the whole system consists of several
series of blocks, namely the BH1750 lux meter sensor as a
receiver of light coming from the front and the MPU6050
accelerometer sensor used to assess the slope of the road. Both
of these sensors function as input, which will be processed by
Arduino Uno. Relay Switching to change the lamp mode from
high beam headlight to low beam or vice versa. A driving servo
to change the direction of the lamp reflector so that the beam
direction can be up or down. Changes in lamp reflectors and
servo motors depend on the process carried out by Arduino
Uno. The headlight on the block diagram as the primary output
in the system.
Automatic headlight system design flowchart using the lux
meter sensor and accelerometer can be seen in Fig. 2. The
input to this system is the angle of the road and the light
of other vehicles. The slope of the road is measured using an
accelerometer sensor. If the road slope measurement is greater
than 5 degrees, the servo motor pushes the top of the reflector
forward 1 degree and the headlight mode changes to low beam
mode. If the slope of the road is less than 5 degrees, the servo
motor pulls the top of the reflector backwards 2 degrees and
the headlight mode changes to low beam mode. The servo
will not move when the tilt angle is zero or pass a flat road.
Measurement of light intensity using a Lux meter. If the light
intensity received by the lux meter is more than 5 lux, the
headlight mode will switch to the low beam mode and then
finished.
IV. R ESULT
Automatic headlight testing process takes place in the
laboratory and actual conditions. Testing in the laboratory to
ensure the system can run well before it is installed on the car.
Research on a laboratory scale is carried out on a prototype of
the front of a car made of wood board material. The prototype
is made with the same dimensions as the actual conditions Fig. 2. Automatic headlight system design flowchart.
so that after laboratory testing, the components and machine
tools can be directly transferred to the car. The prototype
of the front of the vehicle has dimensions with a length of
40 cm; width 23.5 cm; and height 29 cm. Headlight system
in the design of this tool is composed of several electronic

113

Authorized licensed use limited to: Auckland University of Technology. Downloaded on December 20,2020 at 15:55:27 UTC from IEEE Xplore. Restrictions apply.
 2020 2nd International Conference on Industrial Electrical and Electronics (ICIEE)

instruments such as MPU6050, BH1750, Servo Motor. The TABLE II

Headlight system also consists of several mechanical parts L UX M ETER M ODULE T ESTING
such as a lamp driving lever, roll on a flexible bracket so Digital Lux Meter BH1750
No Calibration
that the bracket can be removed, and the bracket that can be Measurement Measurement
moved so that the Headlight reflector can move. 1 Indoor (lights off) 0 0
2 Indoor (lights on) 66 54
In laboratory testing, the MPU6050 module as an ac- 3 5 watt flashlight 141 131
celerometer, the BH1750 module as a lux meter, MG995 Servo 4 5 watt lamp 30 28
Motor movement testing and overall tool testing is tested. 5 Laboratory 35 35
In accelerometer testing, the MPU6050 module is manually
calibrated. Calibration is performed using a ruler bow and
water pass. Arc ruler used to determine the amount of the
angle measured by the sensor and water pass used to ensure
calibration is done in a flat field. Testing is done by rotating
the position of the sensor from the slope of 0 degrees to 90
degrees. So that the values obtained in Table I.

TABLE I
ACCELEROMETER S ENSOR T ESTING

Difference
Trial Raw Measurement
No Y/16384 Angle
Angle Data Y Angle
Measurement
1 0o 132 0,0081 0,4614o 0,4676o Fig. 3. PWM Servo Signal Forms.
2 5o 1540 0,0940 5,3934o 0,3934o
3 10o 2572 0,1570 9,0318o 0,9682o
4 15o 3562 0,2174 12,5568o 2,4432o position, which is 90 degrees. Then the servo motor moves
5 20o 5564 0,3396 19,8525o 0,1475o
6 25o 7016 0,4284 25,3548o 0,3548o
in the direction of 120 degrees when the road is uphill and
7 30o 8024 0,4897 29,3239o 0,6761o in the direction of 70 degrees when the road is downhill.
8 35o 9288 0,5669 34,5340o 0,4660o Oscilloscope signal readings on the servo motor are in Fig. 3.
9 40o 10544 0,6436 40,0574o 0,0574o Testing using an oscilloscope aims to determine the waveforms
10 45o 11520 0,7031 44,6783o 0,3217o
11 50o 12524 0,7644 49,8540o 0,1460o of PWM signals generated by servo motor movements when
12 55o 13280 0,8105 54,1494o 0,8506o the road conditions are uphill or downhill. The shape of the
13 60o 14224 0,8682 60,2460o 0,2460o PWM signal (Fig. 3) has a different pulse width at each angle.
14 65o 14924 0,9109 65,6284o 0,6284o
The pulse width measurement for each test can be seen in
15 70o 15272 0,9321 68,7691o 1,2309o
16 75o 15772 0,9626 74,2904o 0,7094o Table 3.
17 80o 16164 0,9866 80,6000o 0,6000o
18 85o 16216 0,9897 81,7879o 3,2121o TABLE III
19 90o 16363 0,9987 87,0988o 2,9012o S ERVO P ULSE W IDTH M EASUREMENT
Deviation 0,8850o
Servo Motion High Pulse Low Pulse Period
No
Angle (ms) (ms) (ms)
Table I shows the results of testing the accelerometer sensor 1 90o 1,47 18,6 20
2 120o 1,78 18,2 20
up and down. The most significant error is 3.21 degrees at 3 70o 1,27 18,8 20
85-degree angle measurements, and the smallest error is 0.06
degrees at 40-degree angle measurements. The average error
in the sensor measurement is 0.8850 degrees. After calibrating the tool, the next step is testing in the
Measurement of light intensity at the driver’s place using laboratory. Laboratory testing is testing the whole system on
a lux meter sensor. In this research, before the BH 1750 a prototype to determine the level of success of the instrument.
module is used in the system, the module is calibrated first Before the tool is mounted on the car, the device is first
by comparing it to other lux meter measuring devices. Testing installed on a one to one scale prototype. Prototype installation
is done in several different conditions. Tests carried out by is done to ensure the mechanical and electrical systems can
measuring the intensity of light at a place with separate lux. work as expected. Testing is done in two ways. First, the
Table II shows the testing of the lux meter measuring device front of the prototype is raised until it reaches an angle of
experienced a slight difference in the measurement of light 3 degrees, 6 degrees, and 9 degrees, then continued on the
intensity. The measurement value of the lux meter measuring back in the same way. Both prototypes are given a beam of
instrument tends to be higher than the BH1750 measurement. light. The success rate of tool response in each test can be
The next test is testing the movement of the Servo Motor seen in Table IV. Laboratory tests carried out are 20 times
MG995. Servo motor to adjust the direction of the reflector for each condition. The working system of this tool has an
lights up and down. At first, the servo motor is in a normal average success rate of 97.3%. The headlight system that has

114

Authorized licensed use limited to: Auckland University of Technology. Downloaded on December 20,2020 at 15:55:27 UTC from IEEE Xplore. Restrictions apply.
 2020 2nd International Conference on Industrial Electrical and Electronics (ICIEE)

been made successfully changes the light mode when it detects C ONCLUSION
uphill, downhill, and when receiving light from the direction Most accidents occur at night, especially if there is no
before it. street lighting on the road. The misuse of high beam lights
TABLE IV
causes some accidents that occur at night. For that, we need a
L AB T ESTING system that can change the car’s headlight mode automatically.
This research design and create automatic headlight based
Number of Percentage
No Testing Tool Form
Experiments of Success
contours of the road and other headlight lights. This study
1 Front Lifted 3o 20 100 uses the BH1750 module as the main component to detect
2
Back to
3o 20 100 other vehicles facing in front of it. This study also used the
Original Position 1 MPU6050 module to assess road contours. That way, the
3 Rear Lifted 3o 20 100
Back to Headlight not only changes the automatic mode based on the
4 3o 20 100
Original Position 2 presence of other vehicles but can also change when passing
5 Front Lifted 6o 20 100 the uphill or downhill road. Both of these sensors function
Back to
6 6o 20 80 as input, which will be processed by Arduino Uno. Relay
Original Position 1
7 Rear Lifted 6o 20 100 Switching to change the lamp mode from high beam headlight
Back to to low beam or vice versa. A driving servo to change the
8 6o 20 85
Original Position 2
9 Front Lifted 9o 20 100
direction of the lamp reflector so that the beam direction can
Back to be up or down.
10 9o 20 100
Original Position 1 Automatic headlight testing process takes place in the
11 Rear Lifted 9o 20 100 laboratory and actual conditions. In laboratory testing, the
Back to
12 9o 20 100 MPU6050 module as an accelerometer, the BH1750 module
Original Position 2
Beam Lights as a lux meter, MG995 Servo Motor movement testing and
13 12 Lux 20 100
From the Front overall tool testing is tested. Overall laboratory testing has a
success rate of 97%, with 260 trials in 13 conditions. And
on real testing, the automatic headlight can switch modes
TABLE V automatically when going uphill and downhill. The headlight
ACTUAL T ESTING
can change modes automatically when passing through roads
Road Measurement Headlight Response with adequate lighting or when passing other vehicles.
No Testing
Headlight Headlight
Tilt Intensity R EFERENCES
Mode Position
Faced with Not
1 -0,15o 11 Lux Low Beam [1] J. Anwar, “Traffic Safety Education and Information Campaign Activi-
another vehicle Change
ties in Indonesia,” pp. 1–20, 2010.
Flat, Not
2 -0,15o 0 Lux High Beam [2] Plainis, S et al. “Road traffic casualties: understanding the night-time
Dark Change
death toll.” Injury prevention : journal of the International Society
Flat, Not
3 0,15o 6 Lux Low Beam for Child and Adolescent Injury Prevention vol. 12,2 (2006): 125-8.
Bright Change
doi:10.1136/ip.2005.011056
Uphill,
4 7,19o 0 Lux Low Beam Down [3] Synák, František and Rievaj, Vladimı́r and Vrábel, Ján and Darko,
Dark Babic. (2018). Examination of the Vehicle Light Intensity in
Uphill,
5 7,19o 6 Lux Low Beam Down Terms Of Road Traffic Safety: A Case Study. 81. 119-128.
Bright 10.14669/AM.VOL81.ART10.
Downhill, [4] S. Lee, So-Eun; Kim, “The Influence of Smart Car Technologies on
6 -7,37o 0 Lux Low Beam Up
Dark Drivers’ Perceived Control and Attachment,” 2017.
Downhill, [5] H. O. Al-sakran, “Intelligent Traffic Information System Based on
7 -7,37o 5 Lux Low Beam Up
Bright Integration of Internet of Things and Agent Technology,” Int. J. Adv.
Comput. Sci. Appl., vol. 6, no. 2, pp. 37–43, 2015.
[6] J. Huang, Z. Huang and K. Chen, ”Combining low-cost Inertial Mea-
After testing in the laboratory, the next step is to examine surement Unit (IMU) and deep learning algorithm for predicting vehicle
the actual conditions. The real test aims to find out whether attitude,” 2017 IEEE Conference on Dependable and Secure Computing,
Taipei, 2017, pp. 237-239.
the automatic Headlight control system can work according [7] I. Krzysztofik, J. Takosoglu, and Z. Koruba, “Selected methods of
to the design that has been made from the beginning to the control of the scanning and tracking gyroscope system mounted on a
end of the car. Testing is done by considering two conditions. combat vehicle,” Annu. Rev. Control, vol. 0, pp. 1–10, 2016.
[8] Y. S. Chen and S. C. Chiu, “New method of automatic control for vehicle
First, the vehicle passes the road with reduced street lighting Headlights,” Optik (Stuttg)., vol. 157, pp. 718–723, 2018.
levels and is sufficient with flat road conditions. Second, the [9] P. Sevekar, “Night-time Vehicle Detection for Automatic Headlight
car passes the road uphill and downhill with enough and less Beam Control,” vol. 157, no. 7, pp. 8–12, 2017.
[10] S. Eum and H. G. Jung, “Enhancing light blob detection for intelligent
lighting to determine the response of the headlight. The results Headlight control using lane detection,” IEEE Trans. Intell. Transp. Syst.,
of the headlight response are in Table V. Based on the test vol. 14, no. 2, pp. 1003–1011, 2013.
results in Table V, the automatic headlight can switch modes
automatically when going uphill and downhill. The headlight
can change modes automatically when passing through roads
with adequate lighting or when passing other vehicles.

115

Authorized licensed use limited to: Auckland University of Technology. Downloaded on December 20,2020 at 15:55:27 UTC from IEEE Xplore. Restrictions apply.