PlantsCare IoT
End-to-end agricultural IoT, from hardware to cloud: custom ESP32 nodes, MQTT/CoAP, InfluxDB, EfficientNet-B0 image classification and AWS deployment.
What it is
A complete, hardware-to-web agricultural monitoring system, built during my time at the Aragón Institute of Technology (ITA). Not just a dashboard: it includes the sensor-node firmware, the IoT communication layer, the time-series database, AI image classification and the cloud deployment.
What I built
- Three ESP32 node types with custom firmware: water (irrigation/humidity), air (environment) and camera — each with its own dedicated firmware.
- Real IoT communication: MQTT and CoAP between nodes and server.
- Time series in InfluxDB, built for continuous sensor readings.
- Plant-image classification with EfficientNet-B0 (TensorFlow Lite) for early problem detection.
- Monitoring web app with Astro + React and a Node backend.
- Deployed on AWS EC2, with the electronics documented (sensor and microcontroller datasheets in the repo).
Why it matters
This is the project that joins my two worlds: real electronics/IoT (firmware, protocols, physical sensors) and monitoring software (time series, applied AI, web). The same combination I apply today to critical infrastructure at OFITECO.
Stack
ESP32 (C/C++ firmware) · MQTT · CoAP · InfluxDB · TensorFlow Lite (EfficientNet-B0) · Astro · React · Node.js · AWS EC2.
Other work with a related technical focus:
Computer vision on autonomous 4G cameras (OFITECO)
Full-stack service to operate autonomous 4G cameras (solar, no fixed IP) and analyse events with computer vision over distributed critical infrastructure.
DIC 3D VR
Immersive (VR) digital twin of ITA's DIC building: Babylon.js + WebXR, a Blender model and real sensor data from InfluxDB.
xdata-bim — Infrastructure Digital Twin (OFITECO)
Web-based digital-twin platform (BIM + GIS + IoT) for monitoring critical infrastructure in real time. Built solo, from architecture to production.