There are many amazing smart home devices out there, which can control almost every aspect of modern homes. But the smart home of today is still struggling to intelligently adapt to the user's habits without individual setup. With this project we are going to change this.
What it does
Our smart home system collects habits from multiple households and uses this data to infer the smart home setup of the individual (e.g. brightness, color, music, behavior). The more the person adapts his home to his preference, the better our system is able to reflect his likings, i.e. adapt to his individual preferences. This learning and inference step is done inside a cloud app.
How we built it
We used a Raspberry Pi and a Zigbee module to control everyday household devices and gathered measurements from our surroundings using sensors on the Thunderboard Sense2 chip. Furthermore, we used a Rest HTTP Akka (data streaming) backend to process the data.
Challenges we ran into
Dealing with faulty hardware and setting up the network was very challenging.
Accomplishments that we are proud of
- Model and Idea
- Setup and control the hardware
- Good teamwork
What I learned
- Working with different hardware devices (RaspberryPi)