Containerized, automated, and AWS-assisted solution for air quality monitoring at home.

Air quality monitoring. Collect CO2, humidity, temperature data using Sensirion SCD41 sensor and upload the data to AWS RDS database (MySQL engine). This is containerized (Docker) and run on Raspberry Pi B+ model (32bit, ARMv6 CPU). Interactive visualization of data on R Shiny app (dashboard).

Automated depolyment (using CI/CD): on git push, a new Docker image will be generated and saved in GitHub Packages. Any changes made to the Shiny app will automatically be deployed to an AWS EC2 instance (where the app is hosted) and update the web app.

This repository is under development and more features will follow.

Prerequisite: conda Clone the repository to your work directory.

git clone [email protected]:leejheth/homeIoT.git
cd homeIoT

Set up the environment. This will create a new conda environment and install all dependencies in it.

make setup

Activate the newly create environment.

conda activate iot-env

Start measurement and data streaming.

python measure.py

• Install Raspberry Pi OS Lite on micro SD card on your PC and insert the card to Raspberry Pi. In advanced setting, configure WLAN credentials and enable ssh connection.
• Set static IP address on your Raspberry Pi, so it is always accessible via ssh connection with the same private IP address.
• Install Docker

sudo apt-get update
sudo apt-get upgrade
curl -fsSL https://get.docker.com -o get-docker.sh
sudo sh get-docker.sh

Verify if the docker installation is successful using the following command.

docker run hello-world

Pull docker image from Docker Hub.

docker image pull leejih/pi-co2-sensor:3.1

In your current directory, create a folder named conf and a file named 'mysql.txt' to store MySQL-related credentials.

mkdir conf
touch conf/mysql.txt

Run docker container.

docker run --mount type=bind,source="$(pwd)"/conf,target=/home/conf --device=/dev/ttyUSB0 -d here_comes_image_ID

To deploy the Shiny app on the server, e.g. on a AWS EC2 instance, there is a great tutorial here.

The current repository employs a CI/CD pipeline (.github/workflows/cicd.yml) to create a docker image (to be run on Raspberry Pi) and to automatically deploy the Shiny app on AWS EC2.

• SCD4x CO2 sensors
• Amazon RDS