Project: Parking App
What I learned: API Resources
Duration: 8 months
Age: 20
This project focused on improving the efficiency of Virginia Tech Parking. The project was done in collaboration with the Parking App VT club, which has now become a startup! Two task were undertaken in this project, firstly build a sensor to detect occupancy of parking lots, secondly build an app to interface with this data.
Sensor Research and Construction:
It was determined that the most cost effective and reliable solution to detect parking lot occupancy was to use audio detection at the entrances and exits of the parking lot. This solution was determined after careful analysis of current parking solutions on the market and previous research (More information can be found in the powerpoint attached). A sensor was designed and constructed but not fully implemented due to termination of the club project. It was planned that a machine learning model would be constructed to classify the audio data from an array of microphones on the sensor.
App:
An app built with MIT app inventor was built to interface with the collected data. The app located the best parking lot to park based on the users location and parking tag. It then showed the user a map with directions to the parking lot, a plot of the occupancy (not implemented with real data) and a live webcamera of the parking lot if available. Because the system relied on live data colleted from sensors a backend was required to handle the request. The back end was built with google spreadsheets, google forums, and googel scripts. This allowed for a reliable,fast and most importantly free back-end for handling app request.
