Inspiration
We wanted to have an all-in one solution concerning home lighting. With this device it should'nt be necessary to use the light switches anymore. It aims to improve energy efficiency and decrease light pollution with intelligent lighting. Further, lights are a good way to provide feedback and keep you safe in case of bad airquality or even hazards like fire breakout.
What it does
The device keeps the brightness of the room at a user-defined level regardless of enviromental influences. It provides a smooth transition between day and night and keeps energy use at a minimum. With the use of multiple sensor data is detects if the room is used and controls the light accordingly. As a health and safety feature air quality and temperature is monitored, triggering optical alarms if critical values are reached. In an event of danger, emergency contacts e.g. family members will be informed.
How we built it
Using multiple sensors the device can assess the environment and change colour and brightness of the lights. A PD controller regulates the brightness using ambient light measurements of the sensor. To check for the presence of people it analyzes the change in CO2 concentration and motion over a timespan in the room ensuring a correct detection. To detect bad air quality the CO2 and the TVOC concentration are monitored. When critical levels are reached light will flash peridocally.For fire breakouts the temperature is further monitored and lights will flash rapidly.
Challenges we ran into
As most of us are not very experienced in project and team managent, coordinating five people was tough but solveable. Additionally we didn't plan out the whole project in the beginning and changed ideas on the fly. Raspberry Pi stopped working multiple times expected burnout.
Accomplishments that we're proud of
Creating a 'a little' smart device in one weekend
What we learned
Working with people with different backgrounds
What's next for SmartFART
What can be further improved is:
- Tuning parameters for constant brightness and people detection
- Display data and offer more individual control via handy app
- Integrate outdoor data e.g. weather data or data of smart lampposts
Built With
- java
- json
- linux
- python
- raspberry-pi
- thunderboard-sense2
- zigbee
Log in or sign up for Devpost to join the conversation.