Ecommerce websites like Ebay know your browsing habits and what you are interested in. But, they lack an understanding of the environment in your everyday life. We envisioned a platform where the data from ones real everyday life could be used to make ecommerce decisions.
What it does
We used an IoT board to detect variables in someones life. These variables included the noise around someone, the temperature over time in their home, and the amount of light in someone's home. When our device detects these different values, it looks to see if any values are unusual and will automatically find a product that fixes the users problem through the google search api.The app is accessible through alexa
How we built it
Challenges we ran into
We had trouble with different operating voltages with the different sensors. We could switch to either the 3.3v regulator or the 5v voltage regulator. In order to cope with the different operating voltages we had to run our AnalogRead data through some algorithms to change the voltage level from 3.3v to 5v and vice-versa. We didn't have a credit card for the AWS Lambda function even though the function was free.
Accomplishments that we're proud of
We are very proud of being able to get our Alexa Skill running because the AWS Lambda NodeJS function was probably the most challenging task for us. We are also proud of being able to learn the complicated algorithms to switch between voltage/logic levels.
What we learned
We learned that switching between 3.3v and 5v isn't as easy as it sounds. It is not a proportion and you need to use natural logs. We also learned to come prepared with a credit card number on your account for AWS or it doesn't let you use the free tier.
What's next for BuySense
Right now we use the Google Search API, we'd like to make our AI that converts the sensor data into an item more advanced. We also would like to port our hardware hack into cheaper board like a pi zero that has a smaller footprint. We feel like BuySense could really be beneficial to the ecommerce world.