IoT-UrbanPulse 🚦

An intelligent IoT-based urban traffic monitoring and congestion prediction system that leverages computer vision, edge computing, and real-time data analytics to provide comprehensive traffic insights for smart cities.

🎯 Overview

UrbanPulse is a complete IoT ecosystem designed to monitor urban traffic conditions in real-time. The system captures images from ESP32-CAM devices, processes them using machine learning models for vehicle detection and congestion analysis, and provides real-time traffic updates through a mobile application.

🏗️ System Architecture

The project consists of four main components working together:

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│                 │    │                 │    │                 │    │                 │
│   IoT Devices   │───▶│  Edge Gateway   │───▶│  Cloud Backend  │───▶│  Mobile App     │
│   (ESP32-CAM)   │    │   (BLE Client)  │    │  (ML + API)     │    │   (React Native)│
│                 │    │                 │    │                 │    │                 │
└─────────────────┘    └─────────────────┘    └─────────────────┘    └─────────────────┘

📱 Components

🔧 Device (/Device)

• Platform: ESP32-CAM with AI-Thinker module
• Technology: Arduino C++, PlatformIO
• Features:
  • High-quality image capture
  • Bluetooth Low Energy (BLE) communication
  • Authentication system for secure access
  • Power management and camera reset functionality

🌐 Edge Gateway (/Edge)

• Technology: Python with asyncio and BLE
• Features:
  • BLE communication with ESP32 devices
  • Image preprocessing and optimization
  • Local caching and batch processing
  • Multi-device management support

☁️ Cloud Infrastructure (/Cloud)

ML Service (cloud-run-service)

• Technology: Flask, ONNX Runtime, OpenCV
• Features:
  • YOLOv5 vehicle detection model
  • Real-time congestion analysis
  • Accident detection algorithms
  • Google Cloud Storage integration
  • PostgreSQL database connectivity

Backend API (UrbanPulse-backend)

• Technology: Node.js, Express, PostgreSQL
• Features:
  • RESTful API for traffic data
  • Real-time status updates
  • News and alerts management
  • Google Cloud SQL integration

📱 Mobile Application (/UrbanPulse)

• Technology: React Native, Expo
• Features:
  • Real-time traffic status visualization
  • Interactive charts and statistics
  • Traffic news and alerts
  • Multi-city support
  • Responsive design with custom themes

🚀 Getting Started

Prerequisites

• Node.js (v18+)
• Python (3.8+)
• PlatformIO CLI
• Google Cloud SDK
• Expo CLI
• PostgreSQL

Device Setup

1. Hardware Requirements:

   • ESP32-CAM AI-Thinker module
   • MicroSD card (optional)
   • Power supply (5V)

2. Installation:

   cd Device
   pio run -t upload

Edge Gateway Setup

1. Install dependencies:

   cd Edge
   pip install -r requirements.txt

2. Configure environment:

   cp .env.example .env
   # Edit .env with your configuration

3. Run the gateway:

   python main.py

Cloud Deployment

ML Service

1. Setup:

   cd Cloud/cloud-run-service
   pip install -r requirements.txt

2. Deploy to Google Cloud Run:

   gcloud run deploy urbanpulse-ml \
     --source . \
     --platform managed \
     --region europe-southwest1

Backend API

1. Setup:

   cd Cloud/UrbanPulse-backend
   npm install

2. Deploy:

   gcloud run deploy urbanpulse-api \
     --source . \
     --platform managed \
     --region europe-southwest1

Mobile App Setup

1. Install dependencies:

   cd UrbanPulse
   npm install

2. Start development server:

   npm start

3. Run on device:

   npm run android  # For Android
   npm run ios      # For iOS

🔬 Machine Learning Model

The system uses a custom-trained YOLOv5 model for vehicle detection with the following capabilities:

• Vehicle Detection: Cars, trucks, motorcycles
• Congestion Analysis: Real-time traffic density calculation
• Accident Detection: Unusual vehicle patterns and collisions
• Traffic Flow: Movement analysis and predictions

Model Performance

• Classes: Vehicle types (cars, trucks, motorcycles)
• Input: 640x640 RGB images
• Format: ONNX for cross-platform compatibility
• Inference: Real-time processing on CPU/GPU

📊 API Endpoints

Traffic Status

• GET /traffic/status - Current traffic conditions
• GET /traffic/news - Traffic news and alerts
• POST /upload - Image upload and processing

Health Monitoring

• GET /health - Service health check

🛠️ Configuration

Environment Variables

# Cloud ML Service
BUCKET_NAME=your-gcs-bucket
MODEL_PATH=app/yolov5su.onnx
DB_HOST=your-cloud-sql-instance
DB_USER=your-db-user
DB_PASS=your-db-password
DB_NAME=urbanpulse

# Backend API
NODE_ENV=production
DB_HOST=your-cloud-sql-instance
DB_USER=your-db-user
DB_PASS=your-db-password
DB_NAME=urbanpulse

# Mobile App
REACT_APP_BACKEND_URL=https://your-api-domain.com

🏙️ Supported Cities

Currently deployed in:

• Sant Cugat del Vallès (Catalonia, Spain)
• Monitoring locations:
  • AP-7 (Km 230, Km 200)
  • Rotonda Pere I
  • C/ Costa i Llobera

📈 Features

Real-time Monitoring

• ✅ Live traffic status updates
• ✅ Vehicle counting and classification
• ✅ Congestion level analysis (Fluid/Dense/Stopped)
• ✅ Accident detection and alerts

Data Analytics

• ✅ Historical traffic patterns
• ✅ Peak hours analysis
• ✅ Traffic flow predictions
• ✅ Statistical reporting

User Experience

• ✅ Intuitive mobile interface
• ✅ Real-time notifications
• ✅ Interactive charts and graphs
• ✅ Multi-language support

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• University: Universitat Autònoma de Barcelona (UAB)
• Course: IoT Systems and Applications
• YOLOv5: Ultralytics for the base detection model
• Google Cloud: Infrastructure and ML services
• Expo: React Native development platform

📧 Contact

Project Team: JG03dev Institution: Universitat Autònoma de Barcelona (UAB)

---

Made with ❤️ for Smart Cities