Shortest Path Algorithm for Smart Parking System

 




 

Guek, Teng Tze (2018) Shortest Path Algorithm for Smart Parking System. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

[img] Text
GUEK TENG TZE - FULL THESIS.pdf
Restricted to Registered users only

Download (5MB)

Abstract

Aim: This project aims to design a smart parking system using shortest path algorithm with the capability to update users the shortest route to available parking slots in an indoor car park. Research Methodology: The project starts with the prototype fabrication of a mini parking system which includes an Arduino Mega 2560 microcontroller and light dependent resistor sensors are installed at all parking slots as availability indicator. The status of sensors and location of users are sent to a computer and results are computed using shortest path algorithm in MATLAB. The computed result i.e. the shortest route to the parking slot will then be sent to users through a customized Android mobile app via ThingSpeak. For the computation of route, two methods are developed. The first method (thereafter Method OBO) is by scanning all user location in a fixed interval, and computing the distance from each user location to all available parking slots. The nearest parking slot will be assigned to that specific user. The second method (thereafter Method ALLO) is by generating a distance matrix for all possible user’s location and parking slots. Shortest distance sorting will then be carried out to determine the best pair of user location and parking slots. Results and Discussions: First, functionality of the prototype and information exchange between MATLAB app in a server and the Android mobile app via ThingSpeak is tested. The shortest route is sent to all users in interactive graphics. Second, the two methods which are developed to identify the nearest user to the available parking slots are tested and compared based on their effectiveness and execution time. Four different testing cases are conducted excluding the system dependency on hardware and IoT speed. The results shows that Method ALLO is found to be more effective as it generated lesser overlapping paths and it takes shorter execution time in general. Third, the comparison between the performance of Dijkstra’s Algorithm and another shortest path algorithm i.e. Bellman-Ford’s Algorithm is carried out based on their execution time to generate the distance matrix as used in Method ALLO. The comparison shows that Dijkstra’s Algorithm has overall shorter computation time. Conclusions: A functionality tested smart parking system which contains a MATLAB application in computing shortest path algorithm in a server answering the queries from users’ Android mobile app for the nearest route to the available parking slots is developed. Two methods to generate the shortest route to the empty parking slots are designed and tested. Method ALLO which uses distance matrix is found to be more efficient. Lastly, Dijkstra’s Algorithm is found to have shorter computation time in generating the distance matrix applied in Method ALLO. Limitation: The designed testing cases (4 cases) might not cover all aspects to fully test the efficiency and effectiveness of the designed shortest route methods. More sophisticated testing cases are welcome to examine and improve the designed methods. Impact: The apps generated, and the developed methods to find shortest route to the available parking slots based on shortest path algorithms could serve as the foundation of development works of this kind.

Item Type: Final Year Project
Subjects: Technology > Mechanical engineering and machinery
Technology > Electrical engineering. Electronics engineering
Faculties: Faculty of Engineering and Technology > Bachelor of Engineering (Honours) Mechatronic
Depositing User: Library Staff
Date Deposited: 10 Oct 2018 07:59
Last Modified: 10 Oct 2018 07:59
URI: https://eprints.tarc.edu.my/id/eprint/274