• Introduction
• Lessons Learned
• Technology Stack
  • Sensors
  • Actors
  • Software technology

Introduction

Three colleagues and I participated the Hackathon Stuttgart 2016, which means three Days of coding and soldering. For three of us (I included) it was the first hackathon.

Our project was built around the idea, that the trending sleep analysis should be used to not only analyze but also control the sleep. We constructed a tiny bed and added sensors & actors (mostly ThinkerForge devices) to demonstrate possible features. (Fig. 1)

Lessons Learned

I tried to implement real-time charts for a Dashboard with JavaScript, and it was much more difficult than I expected: After a short research, I first tried D3.js, but as soon as the charts got a little bit more complicated, I got visual glitches. It took me a while to find at least one dedicated open source JavaScript library: Project Eon is built on top of c3.js. I still got some visual glitches with it, but it was the best I could find.

Through the search for real-time charting libraries, I also discovered PubNub, one of the supports of Project Eon. PubNub offers real-time services, among those is also a message broker: You can publish data to a channel and consume or subscribe this channel with other applications. The charts from Project Eon accept these channels as data sources. PubNub was surprisingly easy to setup and use.

The Hackathon was so much fun, and looking at the results of all Teams, it really showed me, what a few people can achieve in just 48 hours if they are passionate and in a supporting environment.

My two favorite projects from the other teams:

• “Internet of Dogs”, a sensor network for detecting earthquakes by connecting sensors-collars for dogs.
• “Super Mario Kehrt”, a Super Mario Cart clone which uses a real, driving industrial sweeper as a game controller, to make sweeping more fun.

Technology Stack

Sensors

• Heat sensing camera, to detect body parts sticking out of the blanket
• Accelerometer, for detecting movement in the bed
• Moisture sensor, to monitor body fluids like sweat
• Webcam, for supervising the bed

Actors

• LED Matrix to visualize the heat detected by the camera
• 3 Steps Motors, to control the blanket position
• 1 Servo Motor, to adjust the angle of the bed

Software technology

• Backend: Python, Java, C
• Frontend: Javascript, HTML, CSS, EON Charts
• Cloud-Services: PubNup, Heroku