Improving public parking by using artificial intelligence

Abstract

In this work, a generic solution for parking management is presented. The solution uses artificial intelligence and remote data reading to implement a detection system, a server, a mobile phone application and a back office. With the mobile phone application, users can retrieve information about available parking spaces, specify their preferences and destination on the map, and receive local parking suggestions from the server. The information about the parking space can come from sensors and cameras (in real time), but also from information communicated by other users about the capacity of the park. This information is collected by the server and used to train a neural network that suggests parking spaces. The neural network makes the appropriate predictions and increases the accuracy of the prediction model over time. An example of how the application works can be seen at: https://www.youtube.com/watch?v=G-MdzFcsBWw.