RoadSOS is an AI-powered IoT road safety and emergency response system engineered to drastically reduce emergency response latency during vehicular collisions and road accidents. By uniting embedded sensor hardware, real-time computer vision, GPS geolocation, and automated GSM/Cloud alert pipelines, RoadSOS detects impacts immediately and dispatches critical situational telemetry to trauma centers, emergency services, and designated contacts.
Important
📥 Official Presentation & Project Documentation:
- 📄 Slide Deck (PowerPoint):
ROADSOS (2).pptx(Full animated slide deck with hardware diagrams & performance metrics) - 📑 Technical Project Report (PDF):
RoadSOS.pdf(Detailed engineering specifications & literature review)
| Slide # | Slide Title | Technical Content Highlighted |
|---|---|---|
| 01 | Title & Executive Summary | Project vision, team credentials, and high-level problem statement. |
| 02 | The Problem: Response Latency | Analysis of traffic accident mortality rates caused by delayed EMS dispatch in India. |
| 03 | System Architecture | Hardware-software block diagram combining MPU6050, RPi, OpenCV, and GSM/GPS. |
| 04 | Impact & Crash Detection | Dual-verification pipeline using G-force thresholding and OpenCV visual collision model. |
| 05 | Emergency Dispatch Pipeline | Multi-channel SOS broadcasting via GSM (SMS/Calls) and MQTT cloud endpoints. |
| 06 | Trauma Center Dashboard | Real-time web GIS interface displaying accident locations, severity, and nearest hospital routing. |
| 07 | Field Testing & Metrics | Response latency reduction benchmarks (from 35 mins average down to < 3 mins automated dispatch). |
| 08 | Future Scope & Conclusion | V2X vehicle communication integration and commercial deployment roadmap. |
graph TD
subgraph Vehicle Hardware Unit
A[Impact Accelerometer / MPU6050] --> D[Embedded Processing Unit]
B[Dashcam Video Stream] --> C[YOLO / OpenCV Accident Detector]
C --> D
E[GPS Module] --> D
end
subgraph Emergency Dispatch & IoT Cloud
D -->|Immediate Alert Payload| F[GSM / MQTT Gateway]
F --> G[RoadSOS Central Dispatcher]
G -->|SMS / Automated Calls| H[Emergency Response / Ambulance]
G -->|Live Coordinates & Video Snapshot| I[Hospital Trauma Center Dashboard]
G -->|SMS Notification| J[Emergency Contacts]
end
- ⚡ Zero-Latency Accident Detection: Combines hardware G-force threshold detection (MPU6050 accelerometer/gyroscope) with visual crash verification to prevent false positives.
- 📍 Automated GPS Geolocation Tagging: Fetches real-time latitude and longitude coordinates during impact for pinpoint rescue mapping.
- 📡 Multi-Channel Emergency Dispatching: Transmits SOS alerts over GSM (SMS/Calls) and MQTT/HTTP cloud services to guarantee message delivery even in low-bandwidth network environments.
- 📹 Live Visual Incident Telemetry: Captures pre- and post-collision video snapshots to give emergency responders instant visibility into incident severity.
- 🖥️ Emergency Response Dashboard: Web-based centralized dashboard displaying real-time incident locations, vehicle IDs, and nearest trauma center routing.
- Main Processing Board: Raspberry Pi 4 / ESP32 / Microcontroller Unit
- Sensors: MPU6050 Accelerometer & Gyroscope, Piezo Impact Sensors
- Connectivity: SIM800L / SIM900A GSM/GPRS Module, NEO-6M GPS Module
- Imaging: Pi Camera Module / USB HD Dashcam
- Language: Python 3
- Computer Vision: OpenCV, YOLO / Custom Detection Models
- IoT & Protocols: MQTT, HTTP REST API, Serial Communication (
pyserial) - Dashboard: HTML5, CSS3, JavaScript, Flask
Ensure Python 3.8+ is installed along with required packages:
pip install opencv-python numpy pyserial requestsConnect the GPS and GSM modules to the serial UART pins (/dev/ttyUSB0 or /dev/ttyS0):
python main.pyDistributed under the MIT License. See LICENSE for details.
Developed by A Sharath — ECE Placement Portfolio