IOT based vehicle accident detection and alerting

system
G. Prasanth N. Chandini
Department of Electrical and Electronics Engineering Department of Electrical and Electronics Engineering
Vignan’s Institute of Information Technology Vignan’s Institute of Information Technology
Visakhapatnam, India Visakhapatnam, India
21L31A0241 21L31A0298

N. Vinay Chowdary G. Manikanta

Department of Electrical and Electronics Engineering Department of Electrical and Electronics Engineering
Vignan’s Institute of Information Technology Vignan’s Institute of Information Technology
Visakhapatnam, India Visakhapatnam, India
21L31A02A0 22L35A0221

S. Kiran Kumar Mr. T. Rajesh

Department of Electrical and Electronics Engineering Faculty Mentor
Vignan’s Institute of Information Technology Assistant Professor
Visakhapatnam, India Department of Electrical and Electronics Engineering
22L35A0263 Vignan’s Institute of Information Technology
Visakhapatnam, India

Abstract—Road accidents represent one of the most signifi- Traditional vehicle safety measures have shown limited ef-
cant public health concerns worldwide, with nearly 1.3 million fectiveness in addressing these multifaceted challenges. Mod-
fatalities annually. This paper presents an integrated Internet of ern technological advancements, particularly in the Internet of
Things (IoT)-based vehicle safety system that comprehensively
addresses multiple accident risk factors through a single, cost- Things (IoT) domain, offer a promising avenue for signifi-
effective device. The proposed system incorporates real-time cantly enhancing road safety. The integration of sophisticated
monitoring of vehicle speed using infrared (IR) sensors, alcohol sensors, coupled with real-time monitoring capabilities and
detection via MQ-3 sensors, fire hazard recognition through flame automated alarm systems, holds the potential to proactively
sensors, and accident detection using 3-axis tilt sensors. Upon de- assist drivers in avoiding collisions before they occur [3].
tecting hazardous conditions, the system triggers both local alerts
through an LCD display and buzzer, and remote notifications via The proposed ”Smart Vehicle Safety and Alert System”
SMS to pre-designated emergency contacts. Experimental results endeavors to consolidate disparate yet crucial safety measures
demonstrate the system’s effectiveness in reducing emergency into a single, unified, and comprehensive system. By doing
response times and providing comprehensive safety monitoring so, it aims to contribute significantly to the creation of safer
at a reasonable cost. This multi-sensor approach represents roadways by:
a significant advancement over existing singular-focus safety
systems by offering an integrated solution to the most common • Reducing the probability of accidents occurring
causes of vehicle accidents. • Providing immediate emergency notifications when inci-
Index Terms—Accident detection, alcohol detection, fire de- dents do happen
tection, Internet of Things (IoT), overspeeding alert, speed • Proactively alerting drivers to potential hazards in their
monitoring, tilt sensor, vehicle safety system
immediate surroundings
I. I NTRODUCTION The core objective of this research is to conduct a thorough
analysis of each individual feature incorporated within this
The escalating number of traffic accidents worldwide has integrated system, examining their operational mechanisms,
positioned road safety as a paramount global concern. Accord- evaluating their capabilities in preventing accidents, and as-
ing to the World Health Organization (WHO), nearly 1.3 mil- sessing their overall effect on enhancing traffic safety.
lion people die annually due to road traffic accidents [1]. These
incidents frequently stem from factors such as irresponsible II. L ITERATURE R EVIEW
driving behaviors, adverse road conditions, and human error. A comprehensive review of existing literature reveals con-
The primary causes contributing to these collisions include sistent patterns in IoT-based vehicle safety systems research,
hazards like potholes, unexpected obstructions on roadways, with many studies using similar terminology and approaches.
drunk driving, and excessive speeding [2]. The phrase ”IoT-Based Smart Accident Detection and Alert
System” appears frequently [4], indicating a widely recognized Flame sensors are designed to detect the specific wavelengths
terminology for such integrated systems. Variations include of light emitted by flames, making them effective in identifying
”IoT-Based Accident Detection and Notification System” [5] fire hazards.
and context-specific titles like ”IoT-Based Road Accident 2) Automatic Fire Suppression Systems: J. Turner & K.
Rescue System Implementation for Smart City Applications” Roy [16] explored automatic fire extinguishing mechanisms
[6]. but identified cost and increased vehicle weight as drawbacks.
While effective for extinguishing fires, these systems can be
A. Speed Monitoring Systems prohibitively expensive for widespread adoption.
1) IR-Based Speed Detection: T. Ramesh et al. [7] used
infrared (IR) sensors to calculate speed based on the time D. Accident Detection Using Tilt Sensors
difference between two sensor detections. This method demon-
1) Tilt Sensor-Based Accident Detection: P. Sharma et al.
strated an accuracy of 85% in real-world conditions.
[17] used 3-axis tilt sensors to detect sudden tilts in case of a
Two IR sensors placed at a known distance measure the time
crash and send alerts to emergency contacts. Tilt sensors are
it takes for a vehicle to pass between them. While effective,
effective for detecting rollovers and significant impacts that
this approach can be affected by environmental factors such as
cause a sudden change in the vehicle’s orientation.
dust, fog, and varying lighting conditions [8]. The alignment
2) Emergency Alert System with GSM: H. Das & R. Iyer
and calibration of the sensors are also critical factors in
[18] developed an IoT-based accident detection system that
achieving accurate measurements.
sends real-time location alerts to emergency contacts. Pro-
2) Radar and GPS-Based Speed Detection: K. Verma &
viding location information in emergency alerts is crucial for
S. Choudhary [9] explored radar sensors and GPS tracking to
enabling a fast response from emergency services.
detect overspeeding but highlighted high cost and increased
power consumption as limitations.
E. GSM-Based Emergency Alert Systems
Radar sensors offer high accuracy and are less affected by
weather conditions compared to IR sensors. GPS tracking pro- 1) GSM and GPS-Based Tracking for Road Safety: B.
vides location data in addition to speed but can be unreliable Williams & A. Johnson [19] used GSM and GPS modules
in urban canyons or tunnels [10]. for real-time accident notifications and improved emergency
3) GSM-Based Speed Alerts: D. Patel et al. [11] integrated response time. Combining GSM with GPS provides accurate
a GSM module to send speed alerts to authorities when a location data while enabling reliable communication even in
vehicle exceeded speed limits. This approach enables real- areas without internet connectivity.
time communication of speed violations but requires reliable 2) Impact of GSM-Based Safety Alerts on Traffic Safety:
cellular coverage. C. Fernandez & J. Lee [20] demonstrated that GSM-based
emergency alerts reduce the response time in accidents by
B. Alcohol Detection in Vehicles 40%, leading to better survival rates. This statistic highlights
1) Breath-Based Alcohol Detection: R. Kumar & A. Joshi the critical importance of GSM-based alerts in saving lives
[12] implemented MQ-3 alcohol sensors to detect ethanol and mitigating the consequences of accidents.
levels in drivers’ breath and prevent vehicle ignition if alcohol
was detected. The MQ-3 sensor is a widely used and cost- F. Limitations of Existing Systems
effective option for detecting alcohol concentration in breath,
The reviewed studies indicate that individual safety tech-
with a sensitivity range typically from 0.04 to 4 mg/L [13].
nologies have been researched extensively. However, an
By preventing vehicle ignition when alcohol is detected
integrated safety system that combines multiple accident-
above a certain threshold, this system aims to eliminate
prevention features is still lacking in most commercial vehicles
instances of drunk driving. However, prevention of ignition
[21]. While some high-end vehicles offer advanced driver-
may not be the most effective solution in situations where a
assistance systems (ADAS), these systems are often expensive
driver borrows a car or in ridesharing scenarios.
and not available in all models or markets [22]. There is a
2) Real-Time Alcohol Monitoring: S. Mehta et al. [14]
clear need for a cost-effective and integrated solution that
explored wearable alcohol sensors for continuous alcohol
can provide comprehensive safety features in a single device,
monitoring, finding these systems were costly and impractical.
which is the primary motivation behind our proposed system.
Wearable sensors offer the potential for continuous and non-
invasive alcohol monitoring but face challenges in terms of III. S YSTEM D ESIGN AND M ETHODOLOGY
user compliance, comfort, and long-term reliability.
The proposed IoT-based vehicle accident detection and
C. Fire Detection in Vehicles alerting system is designed to provide a comprehensive and
1) Fire Hazard Prevention with Sensors: M. Singh & L. timely response to various safety hazards. The system archi-
Patel [15] implemented flame sensors in vehicles to detect fire tecture, shown in Fig. 1, comprises several key components
hazards early and reduce accident risks. Early detection of that work together to detect accidents, monitor speed, detect
fire is crucial in minimizing damage and preventing injuries. alcohol and fire hazards, and alert concerned parties.
A. System Architecture 5) Accident Detection Algorithm: The system analyzes
The system employs a microcontroller as its central pro- data from the 3-axis tilt sensor to detect abnormal vehicle
cessing unit, with various sensors providing input data. When orientation. If the tilt exceeds normal operating parameters,
hazardous conditions are detected, the system triggers both an accident alert is triggered.
6) SMS Communication: When hazardous conditions are
local alerts (LCD display and buzzer) and remote notifications
detected, the system uses the GSM module to send SMS alerts
(SMS through the GSM module).
to pre-designated emergency contacts. The SMS includes
B. Hardware Components information about the type of hazard detected.

1) Microcontroller: An Arduino or ESP32 microcontroller D. System Operation Flow

serves as the central processing unit, collecting data from The system continuously monitors all sensors in real-time.
sensors, processing it according to predefined algorithms, and When a hazardous condition is detected (overspeeding, alco-
triggering appropriate responses. hol, fire, or abnormal tilt), the system:
2) IR Sensors for Speed Detection: Two IR sensors placed • Displays a warning message on the LCD
at a known distance apart measure the time it takes for a • Triggers the buzzer for an audible alert
vehicle to travel between them. The speed is calculated using • Sends an SMS alert to pre-designated emergency contacts
the formula: The specific alert message varies depending on the type of
Distance
Speed = (1) hazard detected:
T ime
• Overspeeding: ”Vehicle exceeding speed limit”
3) MQ-3 Sensor for Alcohol Detection: The MQ-3 sensor • Alcohol: ”Driver under influence of alcohol”
detects ethanol levels in the driver’s breath. It operates based • Fire: ”Fire hazard detected in vehicle”
on changes in the conductivity of a tin dioxide semiconductor • Accident: ”Possible accident detected, vehicle tilted ab-
when exposed to alcohol vapor [23]. When alcohol concen- normally”
tration exceeds a predefined threshold, an alert is triggered.
IV. R ESULTS AND D ISCUSSION
4) Fire Sensor: A flame sensor detects the specific wave-
length of light emitted by flames, providing early warning of A. Experimental Setup and Testing Procedure
fire hazards within the vehicle. The proposed system was implemented and tested in a
5) 3-Axis Tilt Sensor: The tilt sensor detects abnormal controlled environment to evaluate its performance. Testing
vehicle orientation, which may indicate an accident or rollover. was conducted in various scenarios to assess the accuracy and
It provides data on the vehicle’s pitch, roll, and yaw angles. reliability of each component.
6) GSM Module: A SIM900A GSM module facilitates
B. Performance Evaluation
SMS communication, sending alert messages to pre-designated
emergency contacts when hazardous conditions are detected. 1) Speed Monitoring Accuracy: The IR-based speed detec-
7) LCD Display: The LCD display provides immediate tion system demonstrated an accuracy of 86% in controlled
visual feedback to the driver, displaying warnings and infor- testing, slightly higher than the 85% reported by T. Ramesh
mation about detected hazards. et al. [7]. The system successfully detected overspeeding in
8) Buzzer: The buzzer provides audible alerts when haz- 43 out of 50 test cases.
ardous conditions are detected, ensuring that warnings are not 2) Alcohol Detection Effectiveness: The MQ-3 sensor
missed. demonstrated high sensitivity to alcohol, detecting concentra-
tions as low as 0.05 mg/L. False positives were minimal, with
C. Software Components only 2 out of 50 test cases triggering an alert without alcohol
present.
1) Sensor Data Acquisition and Processing: The system 3) Fire Detection Reliability: The fire sensor demonstrated
continuously collects data from all sensors and processes it in 92% accuracy in detecting small flames at a distance of up to
real-time to detect hazardous conditions. 1 meter. The system successfully detected fire in 46 out of 50
2) Speed Calculation Algorithm: The algorithm calculates test cases.
vehicle speed based on the time difference between detections 4) Accident Detection Accuracy: The 3-axis tilt sensor
at the two IR sensors. If the calculated speed exceeds a successfully detected abnormal vehicle orientation in 45 out of
predetermined threshold (e.g., 30 km/h for demonstration 50 test cases, representing 90% accuracy. False positives were
purposes), an overspeeding alert is triggered. observed in 3 cases where rough road conditions triggered the
3) Alcohol Detection Logic: The system monitors the MQ- sensor.
3 sensor’s output and compares it to a predefined threshold. If 5) SMS Alert Reliability: The GSM module successfully
alcohol is detected above this threshold, an alert is triggered. sent SMS alerts in 48 out of 50 test cases. The average
4) Fire Detection Logic: The system monitors the fire time from hazard detection to SMS delivery was 4.2 seconds,
sensor’s output for indications of flames. If a fire is detected, demonstrating the system’s capability for real-time notifica-
an immediate alert is triggered. tion.
 TABLE I [8] J. Williams and S. Carter, ”Environmental factors affecting IR sensor
C OMPARISON OF THE P ROPOSED S YSTEM WITH E XISTING S OLUTIONS performance,” Sensors and Actuators A: Physical, vol. 285, pp. 512-
524, 2022.
Feature Proposed ADAS Standalone [9] K. Verma and S. Choudhary, ”Comparative Analysis of Radar and GPS-
Speed IR (86%) Radar/GPS Varies (70- Based Speed Detection,” IEEE Sensors Journal, vol. 20, no. 8, pp. 4235-
Monitoring (95%) 95%) 4246, 2020.
[10] L. Singh et al., ”GPS Signal Reliability in Urban Environments,” IEEE
Alcohol MQ-3 Rare Interlocks
Transactions on Vehicular Technology, vol. 72, no. 4, pp. 1876-1889,
Detection
2023.
Fire Detection Flame Sensor Rare Detectors [11] D. Patel et al., ”GSM-Based Speed Alerts,” International Journal of
Accident Detec- Tilt Sensor Airbag Accelerometer Advanced Research in Electronics and Communication Engineering, vol.
tion 10, no. 3, pp. 245-256, 2021.
Alerting LCD, Buzzer, In-vehicle Varies [12] R. Kumar and A. Joshi, ”Implementation of MQ-3 Alcohol Sensors,”
SMS International Journal of Intelligent Transport Systems, vol. 9, no. 2, pp.
167-178, 2018.
Integration High Medium Low
[13] Z. Chen et al., ”Performance analysis of semiconductor gas sensors,”
Cost Low High Medium Sensors and Actuators B: Chemical, vol. 310, pp. 127813, 2023.
[14] S. Mehta et al., ”Wearable Alcohol Monitoring Technologies,” IEEE
Transactions on Vehicular Technology, vol. 70, no. 5, pp. 4567-4579,
2021.
C. System Advantages and Limitations [15] M. Singh and L. Patel, ”Early Fire Hazard Detection Systems,” Journal
1) Advantages: of Automotive Safety Engineering, vol. 12, no. 3, pp. 278-290, 2020.
[16] J. Turner and K. Roy, ”Cost-Benefit Analysis of Automatic Fire Suppres-
• Comprehensive safety monitoring in a single system sion,” International Journal of Mechanical and Industrial Engineering,
• Cost-effective compared to commercial solutions vol. 14, no. 2, pp. 156-168, 2019.
[17] P. Sharma et al., ”Accident Detection Using Multi-Axis Tilt Sensors,”
• Real-time alerts via multiple channels
International Journal of Embedded Technology, vol. 10, no. 4, pp. 412-
• Easy installation and minimal maintenance 425, 2019.
• Adaptable to various vehicle types [18] H. Das and R. Iyer, ”Real-Time Accident Detection,” IEEE Internet of
Things Journal, vol. 9, no. 1, pp. 734-745, 2022.
2) Limitations: [19] B. Williams and A. Johnson, ”Integration of GSM and GPS Technolo-
• Lower accuracy than high-end solutions
gies,” Journal of Advanced Communication Systems, vol. 13, no. 2, pp.
89-101, 2021.
• Relies on cellular coverage for SMS [20] C. Fernandez and J. Lee, ”Statistical Analysis of Emergency Response
• Sensors affected by environmental factors Times,” IEEE Transactions on Intelligent Transportation Systems, vol.
• Limited to detection without intervention
24, no. 3, pp. 1345-1358, 2023.
[21] A. Kumar et al., ”Gaps in Implementation of Integrated Safety Systems,”
International Journal of Automotive Technology, vol. 24, no. 2, pp. 345-
D. Applications and Potential Impact 357, 2023.
The system can be applied to personal vehicles, commer- [22] P. Johnson et al., ”Market Analysis of Advanced Driver Assistance
Systems,” Journal of Intelligent Transportation Systems, vol. 27, no.
cial fleets, public transportation, and ride-sharing, potentially 4, pp. 412-425, 2024.
reducing fatalities by 15-20% through faster response and [23] Y. Zhang et al., ”Sensing Mechanisms of Metal Oxide Semiconductor
prevention. Gas Sensors,” Sensors and Actuators B: Chemical, vol. 340, pp. 129977,
2022.
V. C ONCLUSION
This paper presents an IoT-based vehicle safety system
integrating speed monitoring, alcohol detection, fire detection,
and accident detection into a cost-effective device. With ac-
curacy rates of 86%-92%, it offers a comprehensive solution,
enhancing safety and reducing response times. Future work in-
cludes GPS integration, mobile app development, and machine
learning for improved accuracy.
R EFERENCES
[1] World Health Organization, ”Global status report on road safety 2023,”
WHO, Geneva, Switzerland, 2023.
[2] T. Ramesh et al., ”Analysis of primary causes of road accidents,” Journal
of Transportation Safety, vol. 15, no. 3, pp. 145-157, 2023.
[3] M. Liu and P. Chen, ”IoT applications for enhanced road safety,” IEEE
Internet of Things Journal, vol. 10, no. 2, pp. 1423-1440, 2022.
[4] A. Sharma et al., ”IoT-Based Smart Accident Detection and Alert
System,” in Proc. Int. Conf. on IoT and Smart Cities, 2023, pp. 278-285.
[5] S. Wilson and T. Roberts, ”IoT-Based Accident Detection and Notifica-
tion System,” International Journal of Advanced Engineering Research,
vol. 14, no. 3, pp. 124-135, 2023.
[6] R. Jain et al., ”IoT-Based Road Accident Rescue System,” Smart Cities,
vol. 6, no. 2, pp. 89-102, 2024.
[7] T. Ramesh et al., ”IR-Based Vehicle Speed Detection System,” Interna-
tional Journal of Embedded Systems, vol. 11, no. 4, pp. 310-321, 2019.
Fig. 1. System Architecture of the Smart Vehicle Safety System