Verdant

Verdant is a vertical indoor farming system that combines IoT, cloud computing, and automation to optimize crop production in urban environments. It integrates real-time monitoring, smart decision-making, and remote control across multiple greenhouses and growing cells.

Results

Control Demo

Monitoring Demo

Features

• Autonomy: Each growing cell operates independently using an ESP8266 equipped with sensors for: temperature, humidity, light, soil moisture, and CO₂; and actuators: water pumps, air pumps, heaters, and LED lighting which allows assuring optimal growing conditions for individual cells using threshold-based control logic.
• Dynamic device registration: MAC address-based ID registration for ESP8266 nodes simplifies deployment.
• Multi-Layer architecture: Cleanly separated into Physical, Edge, Gateway, and Cloud layers for modular design and extensibility.
• Centralized control: Next.js interface for live monitoring, manual overrides, and visual analytics, the front-end is supported by a NestJS backend, the whole application is deployed on Azure Kubernetes Service with Event Grid acting as the central MQTT message broker.

System Architecture

        +-------------------------+
        |       Azure Cloud       |
        |-------------------------|
        | Nest.js APIs + DB (AKS) |
        | Event Grid Messaging    |
        +-------------------------+
                 ↑      ↓
        +-------------------------+
        | Raspberry Pi Gateway    |
        | Mosquitto Broker        |
        | Python Forwarder        |
        +-------------------------+
                 ↑      ↓
        +-------------------------+
        | ESP8266 Node            |
        | ↔ Arduino Uno           |
        | Sensors + Actuators     |
        +-------------------------+

Technologies & Hardware

Microcontrollers

• ESP8266: Wi-Fi node, actuator logic, ID assignment, local decision-making
• Arduino Uno: Sensor interfacing, analog data acquisition, UART to ESP

Microprocessors

• Raspberry Pi: MQTT Broker + Python bridge to cloud

Sensors

• DHT22: Temp/Humidity
• LDR: Ambient Light
• Soil Moisture: Irrigation threshold
• MH-Z19B CO₂: Ventilation control

Actuators

• Air Pump: Ventilation/CO₂ dilution
• Water Pump: Irrigation system
• LED Light: Growth-phase lighting

Software Stack

• Backend: Nest.js, PostgreSQL
• Frontend: Next.js, Charts & Live Data Display
• Cloud: Azure AKS, Event Grid

Validated Capabilities

• MAC-based ESP ID assignment & EEPROM persistence
• Real-time sensor readings and MQTT publication
• Actuator triggering based on thresholds
• Hourly consumption calculations (time × flow rate/power)
• Remote manual control through UI
• System resilience with offline fallback logic
• Centralized cloud analytics for multiple greenhouses

Source code

• Infrastructure
• Backend
• Frontend
• Raspberry Pi
• ESP
• Arduino Uno

Project Team

• MERZOUKA Younes Abdel Samade
• TOUBAL SEGHIR Kheir Eddine
• MOULAY LAKHDAR Zakaria
• FENDI Mohammed
• HARRATH Bouchra

🎓 Higher School of Computer Science Sidi Bel Abbès
🧑‍🏫 Supervised by: Pr. RAHMOUN Abdellatif