Sensor Research with Raspberry Pis

Getting into some neat new research, check in for some new things soon. Getting it off the ground as we speak (or read?)

API Specification

POST / - Add a new entry

Add a new entry to the database. The request body must be a JSON object of the form:

datetime: (string) MySQL parsable datetime object
humidity: (number) Percent humidity
fahrenheit: (number) Temperature in fahrenheit
pressure: (number) Atmospheric pressure (in atm)

//e.g.
{
    'datetime': '2018-11-13 09:25:10',
    'humidity': 40.434,
    'pressure': 1.065,
    'fahrenheit': 72.567
}

All fields are optional, although the data is pretty much useless without the datetime field for the time-being.

GET /entries - Get all entries

Retrieve all of the data that the server currently has loaded. This will send back a .tsv (tab-separated values) file for you to download. This can be opened in just about any spreadsheet application for analysis.

POST /entries - Place a batch of entries

Very similar to the POST / endpoint, but accepts an array of objects to all be placed at once.

//e.g.
[
    {
        'datetime': '2018-11-13 09:25:10',
        'humidity': 40.434,
        'pressure': 1.065,
        'fahrenheit': 72.567
    },
    {
        'datetime': '2018-11-13 09:32:55',
        'humidity': 41.007,
        'pressure': 1.063,
        'fahrenheit': 73.224
    }
    //...
]

Getting Started

If you're new to the project, welcome! There is some stuff you'll need to know -- you can find most of it below. If you need more information, please reach out to me via email -- Professor Vesonder or the People Finder can give you my address.

Raspberry Pi

What is a Raspberry Pi? They are virtually the same as any other computer, just on a much smaller scale. Our Raspberry Pi units are loaded with a special operating system called Raspbian, which is a distribution stemming from Debian Linux (hence the mashup between Raspberry and Debian).

If you're familiar with other Debian versions of Linux, such as Ubuntu or Mint, you'll likely feel right at home in Raspbian. If not, you may want to familiarize yourself with Linux and the apt package manager (from Debian), as there is a fair amount of system administration needed for this project.

Project Specifics

The ultimate goal of the project is to derive useful metrics for the design of smart cities using environmental information gathered from a network of Raspberry Pi sensor units. Each of these Raspberry Pi units is equipped with a BME680 sensor, which can sample the nearby space for items such as temperature, humidity, pressure, etc. We'll need to collect as much data on these features as possible to generate our conclusions.

At a high level, the project involves a network of Raspberry Pi units that all communicate with a central server named Mallard. For our usage, the Raspberry Pi units will act as clients and Mallard will act as the server -- the reverse never occurs. If you're confused, please check out the , as it is very commonly used to describe systems such as this one.

Client Side

Recall that the Raspberry Pi units are outfitted with a BME680 sensor. Thankfully, some folks over at developed a Python package to programmatically interface with it. To keep things simple, everything on the client side is written in either Bash or Python -- two very important languages to know! Python is used to collect and upload our data, while Bash handles much of the system-level stuff like Cron.

Cron ensures that our Python scripts sample the environment in 1 minute intervals, the results of which are appended to a .csv file. Every 4 hours, the local .csv file is emptied and the data is sent over to Mallard. This web request is performed using the requests package for Python.

Server Side

A variety of stuff is going on with Mallard. Up front, we have a Node.js/Express web server that will listen for incoming data from Raspberry Pi units. Upon receipt, this data is dumped into a MySQL database.

To interface with the database, you can interact with the API provided by the web server -- some of this is documented above. For example, a GET request on /entries will download a .tsv file of every stored record. Feel free to check out the source (likely routes.js) for the specifics.

mfreema1/pi-research