Proceedings of the Second International Conference on Electronics and Sustainable Communication Systems (ICESC-2021)

IEEE Xplore Part Number: CFP21V66-ART; ISBN: 978-1-6654-2867-5

Smart Accident Detection System

Murukurthi Lokesh Sai Kumar, Uppalapati Sai Ashritha, Yelagandula Sumanth, Shaik Hafeez, Preetha P.K, Aswathy Chandran
Department of Electrical and Electronics Engineering, Amrita Vishwa Vidyapeetham, Amritapuri, India

Abstract— Road traffic accidents are considered an issue of accident prevention and detection systems. An alcohol sensor
significant concern with ever-growing traffic in the transport has been used to trace the alcohol particles in the air and an
sector. The frequency of road accidents has increased eye sensor to check the driver's drowsiness and alert him to
2021 Second International Conference on Electronics and Sustainable Communication Systems (ICESC) | 978-1-6654-2867-5/21/$31.00 ©2021 IEEE | DOI: 10.1109/ICESC51422.2021.9532702

tremendously in this modern era, resulting in human life loss and prevent the accident [2-4]. Intelligent traffic management
property loss. The main reason for the loss of life is the delay in systems have been deployed to prevent road accidents using li-
alerting the concerned authorities. The time taken for the
fi communication between vehicles and traffic light signals.
response team to reach the accident location increases the
chances of mortality. Recently, as technology is advancing in These systems can manage the traffic and ensure efficient
every area, new systems are being proposed to prevent and detect mobility for emergency vehicles [5-7]. Sensors like
accidents to save lives. The problem is approached by taking accelerometers, crash sensors, proximity, capacity, and
advantage of mobile communication networks and the Internet of vibration sensors have been used to detect accidents. They
Things. The paper proposes a simulation model that detects monitor the speed, sudden change in the direction, or any
accidents using sensors. The circuit with sensors is integrated deformity in the vehicle due to the impact. Integrating these
with Arduino, which is designed and simulated in Proteus sensors with a microcontroller establishes the connection
Simulation Tool. When the accident is detected, a notification between the tracking and communication systems. When the
message is sent to friends and family through the detection app.
accident is detected, the location is known using a GPS
Virtual Serial Port Emulator software connects the Proteus
Simulation Tool and IoT platform. Blynk IoT Platform is used to tracking system. The location is sent to the emergency
design the prototype of the android application to detect authorities and police stations with the help of GSM [8-14].
accidents. Upcoming technologies like IoT paved a better and a
Keywords— Accident Detection; Road Traffic; Internet of sophisticated way to solve the most complex problems. These
Things; Blynk IoT Platform; Virtual Serial Port Emulator. technologies are used to create intelligent detection systems
that can save lives . An accident is detected by sensors and the
I. INTRODUCTION Data is sent to IBM IoT Watson and Bluemix via node-red to
authorities. Web interfaces are also used to create an HTTP
The world is constantly changing, adapting, and evolving request automatically in case of an emergency to alert the
with modern technology. Technological advancements have authorities when there is an internet connection and in case of
made transportation much easier and faster. But a growing no internet, the message is sent through GSM [15, 16].
issue is the increase in road accidents leading to public health Android smart phones have an in-built real-time tracking
issues, loss of human life, and property. An article published application that helps to deploy the medical services in time
by World Health Organization (WHO) in February 2020 states during emergencies [17].
that the mortality rate is approximately 1.35 million due to The status of the victims can be known beforehand with an
road accidents and 20 to 50 million people are disabled with ECG monitoring system and IoT application, which can
non-fatal injuries. More than 90 percent of the accidents occur process and visualize patient ECG signals continuously that
in low and middle-income countries and causing the death of helps the medical services to give the proper first-aid [18].An
children and young adults aged 5-29 years [1]. With such an android application called blynk is used for smart hospital
increase in fatality rate, it is essential to improve road safety in systems in order to monitor the condition of victim. [19, 20].
developed and, most importantly, developing countries. One of the major concerns in the existing systems is the
India is a developing country with a huge population which probability of fault accident detection. Information on the
results in a high frequency of road accidents causing loss of health status of the victim is also very important when alert
lives and property. Many accidents happen because of over- messages are sent. More-over the application developed for
speeding, drunk driving, and violating traffic regulations. The accident status display should be easily available to all. In the
accidents that occur because of over-speeding are fatal as the proposed accident detection system these issues are well taken
impact on the vehicle and on the victims is high. Accidents care of.
that occur in deserted places make it difficult to identify them
and send a rescue team to help. One of the major reasons for II. PROPOSED SYSTEM
the loss of lives due to vehicle accidents is that the victims In the proposed model, an accident detection system is
couldn't get the help they need in time. To save those lives and developed by sensing the impact on the vehicle during the
send the response team to the location as quickly as possible, crash. The severity of the accident can also be detected by the
accident detection and notification systems are being deformation caused by the accident. This detection system is
developed with the latest technologies.Some of the work done connected to mobile communication systems and when an
to reduce the mortality rate by road accidents includes both accident is detected, the notification message is automatically

978-1-6654-2867-5/21/$31.00 ©2021 IEEE 1675

Authorized licensed use limited to: COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY. Downloaded on July 17,2025 at 06:55:57 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Electronics and Sustainable Communication Systems (ICESC-2021)
IEEE Xplore Part Number: CFP21V66-ART; ISBN: 978-1-6654-2867-5

sent. GSM mobile communication is used to send the alert in app and the notification message is sent to them. In case of no
case there is no internet connection. A prototype of the internet connection, the accident notification message is sent
Android application is developed using IoT platforms to the mobile number stored in the Global System for Mobile
which enables users to download the application on the mobile Communications (GSM).
phone and give the details of their family and friends in case
of an emergency through the internet connection. The
detection device is installed in the vehicle and connected with
the android app in the victim’s Smartphone. When the device
detects the accident, the notification is sent to the contacts that
are given the android app. The system detects the accident
only when all the sensors reach their pre-defined threshold
values. The accident is not detected even if a single sensor did
not reach its pre-defined threshold value. The probability of
false accident detection is very less as all the conditions must
be satisfied in order to detect the accident.
A. System Architecture
In the proposed system, sensors have been used to detect
the accident and the status of the victim.
a. Pressure Sensor: The pressure sensor is integrated for the
detection of the crash and calculates the pressure applied on
the car due to the impact. The pressure on the car deforms it
and this applied force is calculated by the sensor. The pressure
sensors can be assembled in the side doors of the vehicle that
can detect side crashes of the vehicle.
b. Ultrasonic Sensor: The basic idea of using the ultrasonic
sensor is to know the deformation caused by the collision.
Ultrasonic sensors are placed on the front and back
windscreens of the vehicle. The distance between the sensors
and bumpers is measured and fixed as a threshold value. In
case of a collision, the distance will be decreased and an
accident is detected.
c. Infrared Sensor: The main objective behind using an IR
sensor is to reduce the probability of false accident detection.
Fig 1. Schematic of the Proposed System
When the vehicle is collided or met with an accident, the
infrared sensor goes HIGH if it detects the human. If the B. Smart-phone Application
sensors do not detect the presence of humans after the Blynk IoT platform is used to develop the prototype of the
collision, the notification is not sent. Accident Detection application. Blynk application is installed
d. GPS Module: The main aim of using a GPS tracker in the in the Smartphone and is connected with the detection device.
proposed system is to know the precise location of the The data from the sensors is transmitted through the
accident by identifying latitudes and longitudes after the internet/wi-fi to the Blynk application. The prototype of the
detection of accident. notification app is developed using Blynk, and it is shared
e. GSM Module: The main reason for Global System for with family and friends. When the sensors detect the accident,
Mobile Communications (GSM) Module in the proposed a notification message is sent to the app as well as to the
solution is to send the precise location of the accident using a shared people.
mobile communication network. The notification message is
sent to the number that is already saved in the GSM module. The advantage of the proposed model is that the probability of
f. Heart rate sensors and temperature sensors are placed in fault accident detection is very less since all the sensors must
the seat belt of the vehicle to know the health status of the reach their designated threshold values for the accident to be
victims. They calculate the heartbeat and temperature of the detected. Only if a passenger is there in the vehicle (as
victims. indicated by the IR sensor), accident detection and alert
notification happens. The heart rate sensor and temperature
Figure 1 gives a brief idea of the working model of the sensor attached to the seat belt can give an idea of the health
proposed system. When an accident occurs, it is detected by status of the victim when the message is sent. In most of the
the sensors in the detection device installed in the vehicle and literature reported, a custom-made app is developed as a
the location is known using the GPS module. The device and mobile interface and the public need to register in the app to
the victim’s smartphone are connected to the internet and have get notifications. But in this work, the Blynk application is
access to the contacts of the family and friends stored in the used, which is already available in the play store. Just by

978-1-6654-2867-5/21/$31.00 ©2021 IEEE 1676

Authorized licensed use limited to: COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY. Downloaded on July 17,2025 at 06:55:57 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Electronics and Sustainable Communication Systems (ICESC-2021)
IEEE Xplore Part Number: CFP21V66-ART; ISBN: 978-1-6654-2867-5

sharing the QR code with family/ friends/authorities, the status

and location of the victim can be identified.

III. RESULTS
Proteus simulation software is used to automate electronic
designs. In the Proteus simulation tool, the sensors are
Fig. 4: Notification Message
integrated with an Arduino board. The code is written in
Arduino IDE, and the hex file is uploaded to the board in
Proteus. Pressure sensors are used to determine pressure Figure 4 shows the notification message with the location
applied on the car due to a crash. It acts as a crash sensor in coordinates sent to the mobile numbers saved in the GSM
vehicles that are helpful to open the airbags. Ultrasonic module when there is no internet/wi-fi connection. The
sensors are used to measure the deformity of the vehicle due to message and email notifications can be sent to multiple
impact, and an Infrared sensor is used to find out the presence mobiles and mail IDs as programmed.
of humans. B. Blynk application
.A. Accident Detection Simulation Accident Detection Application prototype is developed
Fig. 2 shows the accident detection and rescue system through Blynk IoT platform. Blynk platform is downloaded in
simulation in Proteus. The threshold values for the pressure the mobile phone, and widgets are added to the project to read
sensor and ultrasonic sensor are 70kpa and 500cm. The value the sensor values from the device in the Proteus. The project is
for the pressure sensor is arbitrarily chosen according to the added to the prototype of the app created through the platform.
sensor available in the Proteus software. The accident is The accident detection and rescue device is developed in
detected when the pressure applied is greater than or equal to Proteus and is connected to Blynk by connecting the COM
70kpa, and the distance measured by the ultrasonic sensor is ports virtually, as in Fig. 5. To interface Proteus framework
less than 500cm, and the infrared sensor is HIGH. The and Blynk application, the Arduino Code is uploaded in the
location is then sent to the family and friends. Proteus framework and in the Virtual Serial port driver
(COMPIM), the ports COM1 and COM2 are paired. In the
Proteus, the input port of COMPIM is inputted as COM2. In
the command prompt, it is verified that the Proteus
Framework and Blynk application are connected. The android
application developed can be shared with friends and family
by generating a QR code.

Fig. 2. Accident Detection and Rescue System Simulation

In figure 3, virtual terminal 1 is shown with values detected
by the sensors. The pressure is above 70kPa, the distance
detected by ultrasonic is less than 500cm, and infrared is
HIGH. This indicates that an accident has occurred, and a
passenger is there in the vehicle. Once an accident is detected,
the temperature of the passenger and his pulse rate is also
sensed to know his health condition.

Fig. 5 Interfacing of Arduino with Blynk

Figure 6 shows the accident detection status display in the
mobile app. The Blynk application is connected to the
Proteus, and the value from the sensors can be seen in the
widgets. The pressure displayed is 74.2kpa, the distance is 335
cm, and the infrared sensor is also activated. Activation of
infrared sensor indicates the presence of a human in the
accident. Since the pressure is greater than 70 kpa, and
Fig. 3: Sensor values displayed on Virtual Terminal distance is less than 500 cm, and infrared is high, all threshold

978-1-6654-2867-5/21/$31.00 ©2021 IEEE 1677

Authorized licensed use limited to: COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY. Downloaded on July 17,2025 at 06:55:57 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Electronics and Sustainable Communication Systems (ICESC-2021)
IEEE Xplore Part Number: CFP21V66-ART; ISBN: 978-1-6654-2867-5

limits are violated, and the condition is detected as a vehicle making use of machine learning techniques can make the
accident. The sensor values are displayed in the mobile app proposed system better and can be considered for future work.
with the location of the accident, and the alert message is sent The proposed system can be easily implemented in existing
to friends and family. The notification as an email also vehicles and this can save precious lives lost due to vehicle
automatically goes to family/ friends and emergency services. accidents.

V REFERENCES
[1] Global Status report on road safety,World Health
Organization(WHO),June 2018.
[2] Vivek Kinage, Piyush Patil, “IoT Based Intelligent System for
Vehicle Accident Prevention and Detection at Real Time,”
Proceedings of the Third International Conference on I-SMAC (IoT
in Social, Mobile, Analytics and Cloud) (I-SMAC 2019) IEEE
Xplore Part Number: CFP19OSV-ART; ISBN:978-1-7281-4365-1.
[3] Nishanth, Ancy John, “Intelligent Vehicle With Accident Detection,”
International Journal of Advanced Scientific Technologies
,Engineering and Management Sciences, Volume.3,Special
Issue.1,April.2017.
[4] Pughazendi N, Sathishkumar R, Balaji S, Sathyavenkateshwaren S,
Subash Chander S and Surendar V, “Heart Attack and Alcohol
Detection Sensor Monitoring in Smart Transportation System using
Internet of Things,” International Conference on Energy,
Communication, Data Analytics and Soft Computing (ICECDS-
2017).
[5] Mugunthan, S. R. "Concept of Li-Fi on Smart Communication
between Vehicles and Traffic Signals", Journal: Journal of
Ubiquitous Computing and Communication Technologies June 2020,
no. 2 (2020): 59-69.
[6] Alvi, Unaiza, Muazzam A. Khan Khattak, Balawal Shabir, Asad
Waqar Malik, and Sher Ramzan Muhammad, "A Comprehensive
Study on IoT Based Accident Detection Systems for Smart
Vehicles," IEEE Access 8 (2020): 122480-122497.
[7] Shaamili R, Ranjith R, Supriya P, "Intelligent Traffic Light System
for Unhampered Mobility of Emergency Vehicles," Proceedings of
the 2nd International Conference on Intelligent Computing and
Control Systems, ICICCS 2018, Institute of Electrical and
Electronics Engineers Inc., p.360-363 (2019).
Fig. 6 Accident Detection status display in mobile application [8] Shivani Sharma, Shoney Sebastian. “IoT based car accident detection
and notification algorithm for general road accidents,” International
Journal of Electrical and Computer Engineering, Vol/Issue 9(5),
October-2016.
IV. CONCULSION [9] Monika Suresh Katake, Yuvraj M.Patil, “Automatic Ambulance
Internet of Things has made the sensors smarter that can Rescue and Psychological Parameter Monitoring,” IRJET, October-
communicate between themselves and are capable of sending 2016.
or receiving data with less or no human interaction. These [10] Kavya, Geetha C.R, “Accident Detection and Ambulance Rescue
using Raspberrypi,” International Journal of Engineering and
technologies can reduce the mortality rate due to accidents. Techniques” May-June-2016.
In this paper, a smart accident detection system is proposed. [11] S. Iyyappan, V. Nandagopal, “Automatic Accident Detetction And
The accident is detected by sensing the pressure impact and Ambulance Rescue With Intelligent Traffic Light System,”
deformation caused due to the accident. Only if a passenger is International Journal of Advanced Research in Electrical, Electronics
and Instrumentation Engineering Vol. 2, Issue 4, April 2013.12.
there in the vehicle it is reported as an accident. The [12] Md. Syedul Amin, Jubayer Jalil, M. B. I. Reaz, “Accident Detection
probability of fault accident detection is very less since all and Reporting System using GPS, GPRS and GSM Technology,”
sensors should reach the threshold limit for the accident IEEE/OSA/IAPR International Conference on Informatics,
detection to happen. The health of the victim is notified by Electronics and Vision.
[13] Amrith, E. Uma Maheshwari, R.U. Anitha, Devi Mani, D.M. Ajay,
sensing the temperature and pulse rate. “Smart detection of vehicle accidents using object sensors with
In the proposed model, the Blynk IoT platform is used to artificial intelligent systems,” IJRTE, Jan-2019.
develop the prototype of the Accident Detection application. [14] P. Kaliuga Lakshmi, C. Thangamani, “An Efficient Vehicle
Blynk application which is downloadable from the play-store, Accident Detection Using Sensor Technology,” International Journal
of Advanced Research in Computer Engineering and Technology
is installed in the Smartphone where messages need to be (IJARCET) Volume 5 Issue 3, March 2016.
received. The data from the sensors and alert messages are [15] Jorge Zaldivar, Carlos T. Calafate, Juan Carlos Cano, Pietro
transmitted through the internet/wi-fi to the Blynk application. Manzoni, “Providing Accident Detection in Vehicular Networks
This helps in the fast detection of accidents and ensuring that Through OBD-II Devices and Android-based Smartphones,” IEEE.
[16] Arsalan Khan, Farzana Bibi, Muhammad Dilshad, Salman Ahmed,
the help will reach the accident victims without any time Zia Ullah, Haider Ali, “Accident Detection and Smart Rescue System
delay. Incorporation of additional sensors for fault detection using Android Smartphone with Real-Time Location Tracking,”
like smoke sensor, sound sensor and additional services (IJACSA) International Journal of Advanced Computer Science and
Applications, Vol. 9, No. 6, 2018.

978-1-6654-2867-5/21/$31.00 ©2021 IEEE 1678

Authorized licensed use limited to: COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY. Downloaded on July 17,2025 at 06:55:57 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Electronics and Sustainable Communication Systems (ICESC-2021)
IEEE Xplore Part Number: CFP21V66-ART; ISBN: 978-1-6654-2867-5

[17] Dathar Abas Hasan, Ayad Ghany Ismaeel, “Designing ECG

Monitoring Healthcare System Based on Internet of Things Blynk
application,”“Journal of Applied Science and Technology Trends,
Vol. 01, No. 03, pp. 106 –111 (2020)”.
[18] Ali, Farman, Amjad Ali, Muhammad Imran, Rizwan Ali Naqvi,
Muhammad Hameed Siddiqi, and Kyung-Sup Kwak, "Traffic
accident detection and condition analysis based on social networking
data,"Accident Analysis & Prevention 151 (2021): 105973.
[19] A. Irfana Kader Nisha, K. Janani, G. Rathi Devi, S. Meivel,
“BLYNK AND MQTT BASED SMART HOSPITAL SYSTEM,”
International Journal of Disaster Recovery and Business Continuity,
Vol.11, No. 1 (2020), pp. 69-79.
[20] Nor Anum Zuraimi Md Noar, Mahanijah Md Kamal, “The
Development of Smart Flood Monitoring System using Ultrasonic
sensor with Blynk Applications,” Proc. of the 4th IEEE International
Conference on Smart Instrumentation, Measurement and
Applications (ICSIMA).

978-1-6654-2867-5/21/$31.00 ©2021 IEEE 1679

Authorized licensed use limited to: COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY. Downloaded on July 17,2025 at 06:55:57 UTC from IEEE Xplore. Restrictions apply.