Goal Report: Alternate Semester
On Alternate-
IoT-Based Intelligent Accident Detection and Rescue System
Submitted by:
Rishibha Jain
102104021
(B. E., Electrical Engineering)
Under the Guidance of
Faculty Supervisor:
Dr. Mukesh Singh
Professor
EIED, TIET, Patiala
2025
Department of Electrical and Instrumentation Engineering
Thapar Institute of Engineering and Technology, Patiala
(Declared as Deemed-to-be-University u/s 3 of the UGC Act., 1956)
Post Bag No. 32, Patiala – 147004
Punjab (India)
i
TABLE OF CONTENTS
DECLARATION..................................................................................................................... iii
INTRODUCTION ................................................................................................................... iv
OBJECTIVES AND GOALS ................................................................................................ iv
METHODOLOGY............................................................................................................... v-vi
TASK IDENTIFICATION ..................................................................................................... vi
PROJECT PLAN AND TIMELINE .................................................................................... vii
GANTT CHART .................................................................................................... vii
DAILY WORK MANAGEMENT… .............................................................. vii-viii
INTERDEPENDENCIES ................................................................................ viii-ix
EXPECTED OUTCOMES...................................................................................................... ix
MENTOR APPROVAL ........................................................................................................... x
ii
Declaration
I hereby declare that the goal report titled, “IoT-Based Intelligent Accident Detection and
Rescue System”, submitted as a partial requirement for the Alternate Semester course
towards the Bachelor of Engineering degree in Electrical Engineering at Thapar Institute of
Engineering and Technology, Patiala, is an accurate representation of the work I undertook.
This work was carried out under the guidance and supervision of:
[Link] Singh, Professor at Thapar Institute of Engineering and Technology, Patiala
(Punjab).
Place: Patiala, Punjab Rishibha Jain
Date: 26/03/2025 102104021
It is certified that the above statement made by the student is correct to the best of my
knowledge and belief.
Faculty Supervisor:
Dr. Mukesh Singh
Professor
EIED, TIET, Patiala
iii
1. Introduction to the Engineering Problem/Opportunity
A. The Rising Issue of Road Accidents
Road traffic accidents are still a major worldwide problem, resulting in thousands
of deaths and injuries annually. Emergency response delays usually result in
aggravated consequences.
B. The requirement for an Intelligent, IoT-Supported Solution
With advancements in IoT technology, it is now possible to have real-time
monitoring and automated alert mechanisms. Through the integration of sensors
and modules into cars, the system has the ability to issue automatic emergency
alerts.
C. Improving Road Safety through Intelligent Engineering
The project will implement a low-cost, energy-efficient, and scalable IoT-based
accident detection system minimizing human intervention and reducing
emergency response time. The system will utilize microcontrollers like ESP32 or
ESP8266 to provide real-time accident detection.
2. Project Objectives and Goals
A. Instant and Precise Accident Detection: Design an IoT -based system incorporating
sensors such as an accelerometer and a gyroscope that can detect the collision of the
vehicle in an instant, preventing delays in responding.
B. Automated Emergency Reporting and Location Sharing: Develop a system that
will automatically report emergency situations with GPS co-ordinates in real-time to
predetermined contacts and emergency services, which will cut response time.
C. Severity Detection for Proper Assistance: Analyzing crash intensity through sensor
readings to identify the seriousness of an accident, emergency services can allocate
priorities to urgent cases for rapid intervention.
D. Low-Power and Stable System Design: Implement low power consumption for
extended monitoring, enabling the system to be efficient and reliable without routine
maintenance or battery depletion problems.
E. Field Testing and Integration: Test the system in real-world accident situations to
determine its accuracy and effectiveness and investigate possible integration with
vehicle safety standards and insurance policies for broader implementation.
4
3. Methodology
A. Hardware Choice and Integration
a. Choose ESP32/ESP8266 as the communication and processing microcontroller.
b. Integrate sensor accelerometer and gyroscope for accidental detection due to
sudden impact and change in motion.
c. Use a GPS module for current location tracking and a GSM/Wi-Fi module for alert
on emergency.
d. Create a low-power circuit to maintain smooth operation with limited energy
consumption.
B. Accident Detection and Communication Based on IoT
a. Design firmware to interpret sensor data and identify collisions with predetermined
threshold values.
b. Set up an emergency alert system that alerts predefined contacts and rescue teams
of crashes.
c. Set up a manual SOS button for the driver to alert in case of emergencies.
C. Machine Learning for Severity Analysis
a. Gather sensor data from multiple crash simulations to train an ML model.
b. Design a classification system to classify the severity of the accident based on
impact force and vehicle motion.
c. Enable edge computing with the ESP32/ESP8266 for in-real-time estimation of
accident severity.
d. Continuously enhance the ML model with real-life data to minimize errors.
D. System Testing and Optimization
a. Ensure power efficiency via sleep modes and effective data transmit strategies.
b. Perform controlled crash tests for verifying the precision of accident detection.
c. Compare system performance with traditional accident reporting techniques.
5
d. Examine response time, false detection rate, and power efficiency to guarantee
system reliability.
4. Task Identification
A. Hardware Selection and Integration
Choose and integrate ESP32/ESP8266 microcontroller, accelerometer, gyroscope,
GPS, and GSM/Wi-Fi modules for real-time accident detection and emergency
notification.
B. Sensor Calibration and Data Processing
Setup accelerometer and gyroscope sensors to sense impact and motion changes.
Create firmware to process sensor data and identify accident occurrence.
C. IoT-Based Communication Setup
Setup MQTT/HTTP-based cloud communication for real-time accident notifications.
Integrate GPS-based location tracking and emergency notification system.
D. Machine Learning Model Development
Gather crash simulation data and train an ML model to predict accident severity. Run
the model on ESP32/ESP8266 for real-time edge-based severity classification.
E. System Optimization and Power Management
Employ low-power modes and data transmission optimization for energy-efficient
operation.
F. Testing and Validation
Carry out controlled crash tests to ensure accuracy in accident detection. Measure
response time, false detection rate, and system reliability through real-world testing.
G. Report Documentation and Analysis
Collect test data, measure system performance, and report results for future
refinement and alignment with automotive safety regulations.
6
5. Project Plan and Timeline
A. Gantt Chart:
B. Daily Work Management:
FEBRUARY AND MARCH 2025
26/02/2025 Wednesday Underwent mentor allocation and title finalization.
28/02/2025 Friday Received project assignment and a comprehensive overview of
the project scope from the mentor.
01/03/2025 Thursday MST
03/03/2025 Thursday Reviewed and analysed existing literature on accident alert and
rescue using iot.
7
7/03/2025 Friday MST
9/03/2025 Monday Conducted an in-depth study of literature on automatic accident
detection and rescue.
13/03/2025 Wednesday Investigated various microcontrollers, held progress meeting
with mentor.
16/03/2025 Thursday Prepared a connection layout for the rescue based system and the
project requirements.
18/03/2025 Monday Created a PowerPoint presentation to document and track project
progress and literature.
21/03/2025 Tuesday Researched various ml algorithms and looked for the one that
suits our novelty.
23/03/2025 Thursday Explored additional online resources for deeper technical
insights related to the project.
24/01/2025 Friday Initiated practical exercises using the ml trained model and raw
dataset.
27/03/2025 Monday Prepared a goal report and got it reviewed from the mentor.
31/03/2025 Tuesday Continued working for configuring the the rescue sytem and
prepared a block diagram.
C. Explanation for the interdependencies for the various tasks are explained as
following:
Hardware Choice and Sensor Integration: The microcontroller
(ESP32/ESP8266) should be properly integrated with sensors (accelerometer,
gyroscope, GPS, and GSM/Wi-Fi modules) in order to support precise
accident detection and real-time communication.
Sensor Calibration and Development of Machine Learning Model:
Preprocessing of raw sensor data and analysis are necessary before training an
ML model for accident severity classification. Sensor calibration quality has a
direct effect on the accuracy of severity prediction.
IoT-Based Communication and Real-Time Alerts: Accident detection software
should be able to trigger IoT-based alerts consistently, sending GPS location
8
and severity information to emergency contacts and rescue teams without any
delay.
System Optimization and Power Management: Power-efficient consumption
strategies like deep sleep modes need to be adopted to provide uninterrupted
monitoring. This relies on optimized data transmission and low-energy
hardware operation.
Testing, Validation, and Performance Analysis: Its effectiveness relies on field
testing to confirm accuracy of accident detection, response time to
emergencies, and system reliability. The outcomes determine improvements to
the model and software refinement.
6. Expected Outcomes
A. A working accident detection system combining ESP32/ESP8266 with
accelerometer, gyroscope, GPS, and GSM/Wi-Fi modules to monitor in real-time
and create alerts.
B. A trained machine learning model that can study sensor data to classify
accident severity, enhancing emergency response prioritization accuracy.
C. Low-power consumption designs and optimized systems to provide consistent
and reliable detection of accidents without high energy losses.
D. Successful real-world and simulated crash testing of the system for accuracy,
reliability, and efficiency of response.
E. Possible integration with automotive safety standards, insurance policies, and
intelligent transportation systems, adding to improved road safety.
9
10
