We have seen numerous approaches to count the number of people currently present in office spaces: cameras, detecting enabled Bluetooth on personal devices, beacons, WiFi connections etc. None of the approaches are really accurate. We have looked around and saw that while some people are standing or walking around, most of the time people are sitting down and working. That's how we came up an idea with of a smart chair with a button to track occupancy of office spaces.
What it does
The prototype displays a circle indicating, whether the button is pressed or not. It also sends a request to the backend submitting its unique identifier and the state of the button every time it changes. The backend accumulates the data, stores it in a database, gathers statistics and allows to see in real time whether a workplace is occupied or not.
The data can be used in a numerous ways. In a shared working environment it can give insights into how many desks are being used and whether the number of workplaces should be increased or decreased. If a company has numerous meeting rooms e.g. for 3, 5 and 7 people, the data can show how efficiently the rooms are used and whether smaller groups of people occupying larger rooms having no reason to do so. The data can show how many hours per day workers spend at their workspaces and what's their present/away pattern throughout the day.
If workers prefers working standing that can be addressed by placing the sensor under the floor tile.
How we built it
We hooked up a button with a LED to Raspberry Pi board. Using its GPIO pins the board can detect button presses and light the button's built-in LED. The board is running Windows 10 IoT with a Universal Windows Platform app. The backend is developed in Python using Django framework. It runs in the cloud in a Docker container on a DigitalOcean instance.
Challenges we ran into
The main challenge was to come up with an idea since we had the skills to build products and systems but no ideas what to work on at first.
Accomplishments that we're proud of
We have a fully working prototype comprised of hardware, frontend and backend software.
What we learned
We haven't had any prior experience working with Windows 10 IoT and it was an interesting learning experience to work with a new platform.
What's next for ButtOn/Off
The cost of the system can be further reduced by replacing Raspberry Pi board with an RFID tag. The button could open and close antenna loop making the tag respond to requests from the reader or ignore them completely. By placing RFID reader under the floor and placing tags with buttons in chairs the system could easily track whether a chair is occupied or not. It could also track the furniture moving between rooms.