EcoDose is a full-stack, AI-powered platform that empowers farmers and agri-enthusiasts to optimize crop yield using Rhizobium-based biofertilizers. By leveraging real-time soil data, machine learning, and generative AI, EcoDose provides precise dosage recommendations, actionable soil feedback, and a context-aware chatbot assistant.
- AI-powered dosage prediction using Random Forest and Linear Regression models
- AI-based soil analysis: Actionable feedback on your soil data (powered by Google Gemini)
- Modern web interface (React + Vite)
- ESP32 firmware for real-time soil data collection and calibration
- Chatbot assistant for agri-queries (Gemini-powered)
- Responsive, user-friendly design
EcoDose/
├── web-app/
│ ├── backend/ # Flask backend (ML models, API, Gemini integration)
│ └── frontend/ # React + Vite frontend
├── firmware/
│ └── ecodose_esp32/ # ESP32 firmware for soil sensors
├── LICENSE # MIT License
└── README.md # (This file)
- See
web-app/backend/README.mdfor backend details - See
web-app/frontend/README.mdfor frontend details - See
firmware/README.mdfor firmware usage
- Python 3.7+
- See
web-app/backend/README.mdfor setup and API usage
- Node.js & npm
- See
web-app/frontend/README.mdfor setup and usage
- See
firmware/README.mdfor wiring, calibration, and upload instructions
- Input soil data (pH, moisture, NPK) via the web app or ESP32 sensors
- Get dosage recommendation (g/m²) instantly
- See AI-based soil analysis: actionable feedback for soil improvement, powered by Gemini
- Chatbot: Ask agri-questions with context-aware AI support
This project is licensed under the MIT License.
© 2025 EcoDose Project. Built by 1st year engineering students at RVCE.