Toretto

Because family never lets family fall down.

Toretto is a mobile fall-detection system designed for families with elderly members. The app continuously reads accelerometer data from a smartphone and feeds it into a Random Forest classifier trained to distinguish falls from everyday movement. When a fall is detected, every member of the user's family group is immediately alerted so they can respond.

How It Works

1. The phone's accelerometer streams real-time motion data (acceleration magnitude, jerk, orientation changes).
2. Features are extracted from short sliding windows of that data.
3. A pre-trained Random Forest model classifies each window as a fall or normal activity.
4. If a fall is detected, the backend pushes an alert to all registered family members.

Tech Stack

Layer	Technology
Mobile	React Native (Expo, TypeScript)
Backend	Python, FastAPI, Uvicorn
ML Model	Random Forest (scikit-learn)

Project Structure

.
├── mobile/          # React Native / Expo app
│   ├── app/         # Screens and navigation (expo-router)
│   ├── components/  # Reusable UI components
│   ├── constants/   # App-wide constants and config
│   └── hooks/       # Custom React hooks
│
└── backend/         # FastAPI server
    ├── main.py      # Application entry point
    ├── config.py    # Environment and settings
    ├── models/      # Pydantic schemas
    └── routes/      # API route modules

Getting Started

Backend

cd backend
python -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env        # add your configuration
uvicorn main:app --reload

The API will be available at http://localhost:8000. Interactive docs are at /docs.

Mobile

cd mobile
npm install
npx expo start

Scan the QR code with Expo Go or press i / a to open in a simulator.

License

MIT