Inspiration
We were inspired to create Find Space when thinking everyday problems that could have a smart solution. Finding available parking is often times tedious and time consuming. That is why we decided to create Find Space to solve this problem. Find Space not only reduces the time spent looking for parking but also can reduce the amount of car emissions wasted and allow drivers to be more aware of other cars and pedestrians in the parking lot.
What it does
Find Space allows drivers to know which area of the parking lot has available parking which can save time and reduce stress. It uses machine learning and computer vision to identify free parking spots and uses led lights to indicate where available parking is.
How we built it
We integrated the Microsoft Azure platform with a Dragon board to create a machine learning powered parking spot detector. It was coded in Python using REST APIs and Arduino libraries. Challenges we ran into
Some of the challenges that we faced include having trouble linking to azure and controlling LED lights using Dragon board (GPIO Pins in particular).
Accomplishments that we're proud of
We are proud of the fact that we were able to learn machne learning and incorporate it into our project as the way to find free parking spots.
What we learned
We learned how to use machine learning, cloud computing, working with a Dragon board and the importance of planning a project. As well as using source control in case mistakes were made.
What's next for Find Space
We hope to be able to example Find Space to be prototyped on a larger scale, using the range of a parking lot to try real life scenario. The machine learning needs to be exposed to more information and parking lot scenarios to improve accuracy. We also hope to make Find Space more environmentally friendly by using solar power as an alternative energy source.