My meteorological station

My meteorological station is a Web app to manage homemade weather stations built with Raspberry Pi.

This is a project for Google Summer of Code 2017 - Physical Web organization.

Table of content

• Getting started
  • Build your station (Raspberry Pi)
    • Station requirements
    • Connect sensors to GPIO
    • Install and configure Raspbian
  • Server installation
    • Server requirements
    • Start server
• Google assistant and API.AI configuration
• License

Getting started

Build your station (Raspberry Pi)

Station requirements

• Raspberry pi (model 2B or 3B)
• Sensor Adafruit DHT22 (3 pins)
• Sensor Adafruit BMP 180 (4 pins)
• Micro SD card (minimum 8GB)
• Micro usb power adapter
• 7 GPIO cable (female both ends)

Connect sensors to GPIO

This image shows how RPi GPIO pins are labeled.

DHT22 sensor:

• + to 3.3v pin
• out to pin 17
• - to Ground pin

BMP180 sensor:

• VIN to 3.3v pin
• GND to Ground pin
• SLC to pin 3
• SDA to pin 2

The final result can be like this:

Install and configure Raspbian

1. Follow the official steps to install Raspbian OS using NOOBS (the easiest way).

2. Enable SSH and I2C port.

   Go to Preferences > Raspberry Pi Configuration > Interfaces tab > Enable SSH and I2C like the next image.

3. Get and annotate RPi ip typing ifconfig in the terminal.

When the server installation is completed and the application is running, you can add a new station and the station is automatically configured by the server (copying all scripts and enabling services using ssh).

Server installation

Server requirements

Install Node.js 8

curl -sL https://deb.nodesource.com/setup_8.x | sudo -E bash -
sudo apt-get install -y nodejs

Create a firebase project

1. Enable Google sign in

   Go to Authentication > Sign-in merthod > Google > Enable

2. Add database rules

   Go to database > Rules and copy this rules.

    {
    	"rules": {
    		"beacons": {
    			"$uid": {
    				".read": "$uid === auth.uid",
    				".write": "$uid === auth.uid"
    			}
    		},
    		"stations": {
    			".read": true,
    			".write": false,
    			"$uid": {
    				".read": true,
    				".write": "$uid === auth.uid"
    			}
    		},
    		"users": {
    			"$uid": {
    				".read": "$uid === auth.uid",
    				".write": "$uid === auth.uid"
    			}
    		}
   
    	}
    }
   

3. Get client data

   Create a file named firebase-config.json with this structure:

    {
    	"apiKey": "",
    	"authDomain": "",
    	"databaseURL": "",
    	"projectId": "",
    	"storageBucket": "",
    	"messagingSenderId": ""
    }
   

   Go to Project Settings > Add Firebase to your web app button and fill all fields with the data from your project.

4. Get admin data

   Go to Project Settings > Service accounts > Generate new private key

   Download and rename key to firebase-admin.json.

Get a maps api key

Create a Google Cloud project, go to APIs & services, enable Maps Javascript API and Maps Geocoding API and create a key.

Create a file named maps-key.json with this structure and paste your key.

{
	"apikey": ""
}

Create a twitter app

Create a file called twitter-config.json with this structure:

{
	"consumer_key": "",
	"consumer_secret": "",
	"access_token": "",
	"access_token_secret": ""
}

Create a new twitter app, go to Keys and Access Tokens and fill all fields with the values from your project.

Start server

1. Clone this repository

    git clone https://github.com/gerard87/my-meteorological-station.git
   

2. Copy config files

   Copy to the root of the project the four config files:

   • firebase-config.json
   • firebase-admin.json
   • maps-key.json
   • twitter-config.json

3. Install dependencies

    npm install
   

4. Start server

    npm start
   

Google assistant and API.AI configuration

1. Go to https://api.ai/ and create a new agent.

2. Click settings > Export and import > Restore from zip.

3. Select the mmsagent.zip inside /assistant/agent folder.

4. Go to Fullfitment and enter the URL: https://<your_server_URL>/assistant

License

MIT © Gerard Farré Gomez