AIRS - An Air Quality Awareness Platform for Smart Cites

Check air quality levels in your city before going for a run in convenient ways such as IFTTT, Amazon Alexa, Twitter, and alerts.

AdvancedShowcase (no instructions)8 hours1,667

AIRS - An Air Quality Awareness Platform for Smart Cites

Things used in this project

Hardware components

Seeed Studio Grove starter kit plus for Intel Edison

Our Grove Air quality sensor is plugged into the Grove shield which is plugged into the Intel Edison board.

Amazon Alexa Amazon Echo

Amazon's Alexa voice assistant was used to read out sensor based, air pollution status

Software apps and online services

IBM BlueMix

We used the pre-configured (boilerplate) IoT solution (find it in the catalog section). This automatically provisioned an IoT stream and node-red instance.

Amazon Alexa Alexa Voice Service

Originally we wanted to have Alexa read directly air pollution status from the Cloudant DB (JSON format) in BlueMix. This wasn't possible in the time that we had unfortunately so as a plan B, we have Alexa read the most recent Tweet from our air pollution Twitter feed.

IFTTT Maker service

IFTTT let's you create your own channel! Search for the "Maker" channel on IFTTT to make your own. You simply do an HTTP POST in this format: https://maker.ifttt.com/trigger/{your_custom_event_name}/with/<your_API_key> With an optional JSON body of: { "value1" : "", "value2" : "", "value3" : "" } Be aware that this custom Maker Channel recipe is only available to your own personal IFTTT login and is not published to all IFTTT users. Nonetheless, it is an excellent to prototype an IFTTT channel super fast. I suppose they probably have a process to take a custom channel and make it an official, global one.

Microsoft Azure

Microsoft Azure IoT was used for the map overlay feature. Azure IoT has a pre-configured remote monitoring solution which turns out to be quite easy to setup. It's good if you need to get something up and running fast. So easy though it was a challenge for us to customize it as we didn't understand the underlying plumbing very well. As with the BlueMix Cloud integration, a small chunk of code is running locally on the Intel Edison to stream data to Azure.

Twitter

We used Twitter as a feed for air pollution status, feed here https://twitter.com/sjstate_airq

Story

AIRS Overview and Demo

We are developing a pollution awareness platform for the smart city that provides interactive environmental information for the residents enabling enhanced well-being. Air pollution is a growing problem causing several health issues including respiratory diseases and cancer. Air quality information is currently not easily accessible. Our product leverages air pollution data from pollution sensors places on lampposts throughout the city. Our solution provides holistic pollution information to the city and residents through a hosted Bluemix-cloud platform. To the user, this will be the first interactive hands-free voice-activated “Amazon Alexa” for Smart Cities. In addition, we have enabled user-definable alerts via IFTTT integration. Finally, this is enabled by a high performance, low cost hardware leveraging Intel Edison.

Our team of 5 worked on this project as part of the Semicon/West Industrial IoT Hackathon over 4 days.

Architecture Diagram

AIRS Architecture Diagram

References:

1. https://github.com/rvanderwerf/alexa-twitter-groovy

2. https://developer.ibm.com/recipes/tutorials/ui-dashboard-for-iot-device-data-using-node-red/

3. https://azure.microsoft.com/en-us/documentation/samples/iot-hub-node-intel-edison-getstartedkit/

4. https://github.com/rwaldron/johnny-five/blob/master/docs/grove-gas-tp401-edison.md

5. https://github.com/Azure-Samples/iot-hub-node-intel-edison-getstartedkit

Schematics

Node-Red Bluemix

Our first time using Node-Red, it's a drag and drop style language to stitch together business logic. A small piece of javascript code is running on the Intel Edison to stream data to BlueMix. The "IBM IoT App In" is connected to that data stream and then now it's a matter of drag and dropping various function blocks to perform your business logic. For example, we had a requirement to store this data for historical analytics. The last box "Cloudant" is a drag and drop event and then specifying a few parameters inside the box we are just like that storing our data in a DB (pretty nice!). BlueMix also has some very powerful capabilities with Watson exposed here which unfrotunately we didn't get time to experiment with during the Hackathon..