Скачать или смотреть SmartSoil: Automated Soil Monitoring & Alert System Demonstration

This video demonstrates the SmartSoil system, an IoT-based soil monitoring and alert platform developed as part of the COM6017M university assignment. The system integrates multiple environmental sensors with edge AI processing to monitor soil and ambient conditions in real time.

Sensor data including soil moisture, temperature, humidity, light intensity, pH level, and a secondary soil sensor is collected using an Arduino UNO R4 WiFi and transmitted via Bluetooth Low Energy (BLE) to a Raspberry Pi 4. The Raspberry Pi performs edge AI inference using a TensorFlow Lite model to classify soil conditions as GOOD, MODERATE, or BAD.

Processed data is uploaded to the ThingSpeak cloud platform, where MATLAB analytics perform trend smoothing and alert evaluation. Based on predefined thresholds and AI predictions, the system generates event-driven alerts. Telegram notifications are sent only when alert states change (OK, Warning, or Danger), implementing an anti-spam mechanism to ensure meaningful user notifications.

The video demonstrates:

Real-time sensor data acquisition
Edge AI soil condition prediction
ThingSpeak dashboard visualisation
Alert logic execution and Telegram notifications
Live soil interaction to validate system responsiveness
All source code, system diagrams, and documentation used in this project are available in the public GitHub repository:

GitHub Repository:
https://github.com/ravisah12345/SmartSoil-...