Smart vehicle parking system based on internet of things

Abstract

As the number of vehicles continues to increase in Malta, efforts are being made on all fronts in order to help curb the negative effects that traffic and parking issues bring about with respect to urban planning, the environment and the economy. The issue of street parking in Malta is becoming more and more of an issue as car density continues to increase, while parking places do not, as well as the lack of regulation from authorities through the use of say, parking meters. The proposed IoT based solution hopes to facilitate the process of parking, and thus decrease the time spent looking for parking and road congestion. In order to provide a strong solution, a popular platform was required which already has a strong user base, as well as meet the requirements for the system to be functional. Hence after defining the needs of the system, mobile phones were decided to be the platform of choice. An application was developed that using GPS and Accelerometers on these phones, would automatically detect when a user has just parked their car, or just left a parking place with their car. With this information, a track could be kept of the number of parking places being taken or created in a given area. All that remained was to distribute this data efficiently and effectively to the user requiring it. Hence a simple UI was create that expects a target on a map, in-putted by the user, and once the user arrives within 2km of the defined target, information about parking places would be displayed on the map for the user to perceive as required. The map displays the parking by putting dots on the coordinates located by the server. Each parking is aged according to how much time has passed since it has been created. The application is crowd-sourced, meaning that it although the system functions, it is still unusable without a number of users using it consistently. After heavy testing, the resulting application manages to detect adding and removing of parking places 100% of the time, assuming proper usage of the system (e.g. moving the phone around in such a way that a step is detected wrongly, might result in a false parking event). That said, locations of parking spots have an average of 6m error. It also contains a fully functional UI which the user could use to interface with the functionality mentioned before. With this in mind, the system was distributed to university students in order to be used mainly for trips to and from university. Multi-platform test have also been performed, as well as sustainability tests. These tests resulted in a number of changes to be implemented as diverse hardware incurred problems that had not materialised on the primary development device. These problems centred around location accuracy and battery consumption. As is, while the system is not perfect, it is proven to be functional, and given more widespread use, could help to point users in the right direction, when looking for street parking places in highly traffic congested areas.