Automobile Black Box System for Accident

Analysis
Monisha J Prasad, Arundathi S, Nayana Anil, Kariyappa B.S
Harshikha Department of Electronics and Communication
Department of Electronics and Communication R V College of Engineering
R V College of Engineering Bangalore, India
Bangalore, India [email protected]
[email protected]

Abstract– Automobiles and computing technologies are The Automobile Black Box exists partially in the
creating a new level of data services in vehicles. The form of Event Data Recorders (EDRs) and black
Automobile Black Box has functions similar to an boxes [4][5][6][7]. It is difficult to establish the exact
airplane black box. It is used to analyze the cause of cause of an accident from these [1]. To overcome this
vehicular accidents and prevent the loss of life and
shortcoming, an Automobile Black Box system that
property arising from vehicle accidents. This paper
proposes a prototype of an Automobile Black Box combines both of these aspects is proposed.
System that can be installed into vehicles. The system
aims to achieve accident analysis by objectively tracking Apart from the accident analysis by objectively
what occurs in vehicles. The system also involves tracking what occurs in vehicles, the proposed
enhancement of security by preventing tampering of the system sends short message indicating the position
Black Box data. In addition, the Black Box sends an of vehicle by GPS system to family member,
alert message to a pre-stored mobile number via Short emergency medical service (EMS) and nearest
Message Service (SMS) in the case of occurrence of an hospital so that first aid can be provided as early as
accident.
possible [8].
The proposed system makes use of 12 sensors to record
the various driving data parameters. The Raspberry Pi The proposed system also incorporates a security
controller (RPi) and Arduino controllers are used to module, which employs data encryption to secure the
regulate these sensors. The data received from the stored data on the SD card [9]. The system aims to
sensors are stored on the SD card mounted on RPi for achieve accident analysis by strategically placed
retrieval after the accident. The system uses external sensors, in and around the vehicle.
sensors such as camera and Global Positioning System
(GPS) to collect video and location data.
II. OVERVIEW OF THE SYSTEM
Keywords – Black Box, Microcontroller, Global
Positioning System (GPS), Global System for Mobile The schematic of the Automobile Black Box System
Communication (GSM) is shown in Figure 1. The output of the sensors is
read from Controller 1 and communicated to
I. INTRODUCTION Controller 2. The data is stored on the SD card. To
prevent falsification of the data, encryption algorithm
According to the World Health Organization (WHO), is used and the data is retrieved using the decryption
more than a million people in the world are losing algorithm.
their life each year because of transportation-related
accidents [1]. To address this problem black box
takes on the role of the investigator to determine the
cause, thereby helping to propose measures to protect
lives. The system also assists the insurance
companies with their claim settlements [2]. The
importance of such system drove the European Union
(EU) to establish a policy that makes it mandatory for
all vehicles to be equipped with black boxes [3].

Figure 1: Schematic of Automobile Black Box System

   !" ##$%&
'


 The Analog to Digital Convertor (ADC) in Arduino
The Block diagram of the Automobile Black Box Uno controller is used to convert analog sensor
System is shown in Figure 2. Raspberry Pi and values to digital values. The analog sensors used are
Arduino controllers are the main components of the accelerometer, alcohol sensor, LDRs. The output
system. They regulate the various sensors interfaced from digital pins of Arduino interfaced to the various
to them. sensor readings are communicated to Raspberry Pi
controller using USB. Raspberry Pi stores these
values and saves it as a .txt file in a Secured Digital
(SD) card which can be read after the accident to
determine the cause of the crash. The GSM module is
interfaced to Raspberry Pi controller using USB
communication. It is used to send the GPS data and
an alert message to pre-stored numbers in the
Subscriber Identity Number (SIM) card. The wheel
speed module is connected to the Raspberry Pi using
the Universal Asynchronous Receiver and
Transmitter (UART) module to record the rpm
(rotations per minute).

Figure 2: Block diagram of the Automobile Black Box System A Real Time Clock is interfaced to the Raspberry Pi
controller using Inter-Integrated Circuit (I2C)
The acceleration of the vehicle at the time of crash is protocol. Real Time Clock is used to record the time
measured using an accelerometer ADXL335. It is a at which the accident occurs. A Raspberry Pi Camera
3-axis MEMS accelerometer measuring static and is interfaced to the Raspberry Pi by plugging into the
dynamic acceleration. The important lights in a Camera Serial Interface (CSI) slot in Raspberry Pi. It
vehicle are the flashers, the brake lights, headlights is used to take video inside the vehicle and store it on
and the rear lights. Light Dependent Resistors an SD card. It can be retrieved later to determine the
(LDRs) are used to detect whether these lights were cause of an accident.
functioning properly during the crash. A wheel speed
module is used to determine the rotations per minute The stored sensor data on the SD card is secured
(rpm) of the vehicle and hence the speed of the using a substitution cipher. The cipher key is used to
vehicle at the time of crash. It consists of Reed decrypt the data stored. The crash sensor similar to
Switches and magnets. The magnets are attached to the one used in air bag deployment is used to sense
the rear wheels of the vehicle. The engine the occurrence of an accident.
temperature of the vehicle is measured using a
temperature sensor DS18B20. A push button is used
to determine whether the driver wore the seat belt at III. METHODOLOGY ADOPTED
the time of crash.
The Arduino Integrated Development Environment
Various warnings are given to the driver using a (IDE) is utilized for programming the Arduino board.
range of sensors. An alcohol sensor module Arduino programs are written in C language. The
consisting of MQ3 gas sensor is used to warn the setup () function is made use of at the start of the
driver when he/she is high on alcohol. An Ultrasonic program to initialize settings like the pinmode of the
Ranging Module HC-SR04 is used for distance digital pins, serial baud rate. The loop() function runs
gauge. The driver is provided warning if the vehicle forever and it has the functions to monitor the various
approaches too close to the vehicle in front or rear. sensors such as alcohol sensor, temperature sensor,
Infra red object sensor modules are used for lane light sensor, accelerometer, ultrasonic sensor, seat
detection. The lane marking of white color on a dark belt, IR sensor, GPS. The serial port is made use of to
road is assumed; when the driver switches the lane, communicate the sensor readings to the Raspberry Pi
the sensor will sense the white line and indicates for storage.
through LED. A GPS system is used to determine the
co-ordinates of the accident location in order to send Python language is used for programming the
it along with an alert message to pre-stored numbers Raspberry Pi board. Threading is used in python
using a GSM module. The GPS module is interfaced programming to run Data Recording and Video
to Arduino using UART protocol. Recording functions.
The flowchart of the overall system is shown in both the threads and the recording stops. The data
Figure 3. stored is encrypted to prevent tampering.

IV. EXPERIMENTAL RESULTS

The different sensors interface to the system is shown

in Figure 4. The complete car model is shown in
Figure 5.

NO

YES

Figure 4: Sensors interfaced to a model car

Figure 3: Flowchart of the system

Threading is used to switch the CPU time between

data recording and video recording. Accordingly, two
threads are defined, Thread 1 to record data
parameters and Thread 2 to record video. When an
accident occurs, the time of crash is noted and the
GPS co-ordinates along with help message are sent
via GSM module. The occurrence of an accident ends Figure 5: Car Model
The values from Arduino are communicated to
Raspberry Pi over USB. The Raspberry Pi virtual
desktop output is shown in Figure 6.

Figure 6: Raspberry pi virtual desktop output

The alcohol value is 167; this is a digital value

indicating the presence of low alcohol, which gives
the information that the driver was not drunk. The
alcohol sensor gives an analog output, which is
connected to the ADC on the Arduino board. The
ADC is 10-bit with reference voltage 5V. Hence an
Figure 7: Text message received through GSM
output of 167 corresponds to voltage level of 0.82V.
The distance of the obstacle in front of the vehicle,
given by the ultrasonic sensor is 8cm indicating the The wheel speed sensor module calculated the rpm of
vehicle is too close. The seat belt status indicates that the rear wheels was shown in Figure 8.
the driver did not wear the seat belt.

The LDR value is a digital output of ADC to which

it is connected. The corresponding voltage is 4.44V
indicating high resistance and that the light is off.
The accelerometer is kept flat on the table, the values
from the pins X, Y correspond to 0g values for the X
and Y direction and the value from the Z pin gives
the 1g value for the Z direction. The process is
repeated in different orientations to get the values of
0g along Z direction, 1g along X and Y directions.
The 0g value and 1g value is subtracted for each
direction and suitable scale is selected. The sensor
values from the pins X, Y, Z are read. These values Figure 8: Wheel speed sensor module output
are subtracted from the corresponding 0g values and
divided by the scale to get the acceleration values. In an effort to incapacitate any manhandling or
tampering of black box data in a crash event, all
The GPS co-ordinates give the location of the information stored in the memory will be encrypted
vehicle. The GSM module communicated the co- by the Raspberry Pi as soon as a crash is detected.
ordinates of the location and a help message to the The encrypted file output is shown in Figure 9.
pre-stored mobile number as shown in Figure 7.
 The material used to protect the car battery can be
made use of for the casing.

VII. REFERENCES

[1] Abdallah Kassem, Rabih Jabr, Ghady Salamouni, Ziad

Khairallah Maalouf, “Vehicle Black Box System”,
Proceedings of the 2nd Annual IEEE System
Conference, IEEE 2008, pp. 1-6.
[2] Byung Yun Lee, Yong Yoon Shin, Hyun Joo Bae, “
Development of Insurance Server System based on
Vehicle Driving Information “, Proceedings of 7th
IEEE International Conference on Computing and
Convergence Technology (ICCCT), 2012, pp. 156 –
159.
[3] Sung-Hyun Baek, Hwa-Sun Kim, Da-Woon Jeong, Mi-
Jin Kim, You-Sin Park, Jong-Wook Jang,
“Implementation Vehicle Driving State System with
Figure 9: Encrypted File OBD-II, MOST network”, Proceedings of the 17th
Asia-Pacific Conference on Communications (APCC),
IEEE 2011, pp. 709 – 714.
[4] Chien-Chuan Lin and Ming-Shi Wang, Department of
V. CONCLUSION
Engineering Science, National Cheng Kung University,
Tainan, Taiwan, “An Implementation of a Vehicular
A prototype of the Automobile Black Box System Digital Video Recorder System”, Proceedings of 2010
was designed and implemented successfully. The IEEE/ACM International Conference on Green
Computing and Communications & 2010 IEEE/ACM
designed system comprises of 12 sensors, which were
International Conference on Cyber, Physical and
placed in and around the automobile. Each of the Social Computing, pp. 907 – 91.
sensors was tested and found to give desired output. [5] Milind Khanapurkar, Dr. Preeti Bajaj, Dakshata
These outputs were communicated to the Arduino Gharode, “A Design Approach for Intelligent Vehicle
Black Box System with Intra-vehicular communication
controller. The Raspberry Pi and Arduino controllers
using LIN/Flex-ray Protocols”, Proceedings of 2008
communicate with each other and regulate the IEEE International Conference on Industrial
sensors successfully. The data retrieved from the Technology IEEE 2008, pp. 1 – 6.
sensors are stored on the SD card successfully and [6] Duy Le Nguyen, Myung-Eui Lee, and Artem Lensky,”
The Design and Implementation of New Vehicle Black
can be fully retrieved when required.
Box Using The OBD Information”, Proceedings of the
7th IEEE International Conference on Computing and
The data is successfully recorded from the video Convergence Technology (ICCCT), IEEE 2012, pp.
camera and is stored as a file on the SD card. The 1281 – 1284.
[7] XuHui, Li Jing-zhao and Yin Zhi-Xiang, Sun Xia,
designed system also incorporates an emergency help
“Design of Vehicle Black Box based on Dual-core
module which automatically alerts the medical System and ȝC/OS-II “, Proceedings of the
services, police and relatives about the accident along International Conference on Industrial Control and
with the GPS co-ordinates. It also has a security Electronics Engineering, IEEE 2012, pp. 763 – 766.
[8] Watthanawisuth, N., Lomas, T., and Tuantranont, A.,
module which will encrypt the data stored to avoid
“Wireless Black Box Using MEMS Accelerometer and
tampering. A help SMS is sent to the pre-stored GPS Tracking for Accidental Monitoring of Vehicles”,
number as soon as a crash is detected. The encryption Proceedings of IEEE-EMBS International Conference
is successfully carried out and decryption of the on Biomedical and Health Informatics (BHI), IEEE
2012, pp. 847 – 850.
encrypted file is accomplished to retrieve the original
[9] Mooseop Kim and Chi Yoon Jeong, “An Efficient Data
file. Integrity Scheme for Preventing Falsification of Car
Black Box”, Proceedings of International Conference
VI. FUTURE SCOPE on ICT Convergence (ICTC), IEEE 2013, pp. 1020 –
1021.
.
The system can be improved by including a face
recognition algorithm focusing on the eye to
continuously monitor consciousness of the driver.
The designed system is implemented in a modeled
vehicle; the same can be interfaced to a real-time
vehicle to gather real-time data. The system can be
made crash proof by providing casing to the sensors
so that the impact is less when an accident occurs.