You are on page 1of 5

A Smart Car Parking System Based on Genetic

Algorithm
Prof. Varsha Bodade, Nidhi Bhat, Jyoti Singh, Ayush Pandita, Shubham Mishra
Department of Information Technology, Mumbai University
Plot no 12, Sector - 22, Phase 2, Nerul(W), Navi Mumbai - 400 706, Maharashtra, India.
bhatnidhi0@gmail.com
jyotimrjs9705@gmail.com
ayushpandita27@gmail.com
mishrashubham2288@gmail.com

Abstract - The information age focuses more on creating a 2) More awareness about situations;
knowledge based society surrounded by high-tech smart
devices. The next trend of innovations is likely to come 3) Analysis of decisions depending upon Geofencing;
from humans' ability to connect to machines and the data
that comes from this connection. The basic problem faced 4) Immediate handling and response generating. Etc
in the present era is the difficulty of finding free parking
slots. The study aims at reducing the waiting time by
providing the real time information about all the parking
More than half of the world’s population is living in
farms available. The parking farm closest to the current cities. So the cities have reached full of their
location is found using Genetic algorithm along with the occupancy. More and more people tend to have their
help of Global Positioning System (GPS) and Geofencing. own personal vehicles which has resulted a large
The application allows pre-booking of parking slots. A number of vehicles further increasing the traffic as well
platform is provided which enables people to cultivate as parking problem. After entering to a new area which
their land into parking areas over a negotiable rent. This is not known already, people spend most of their time
will not only decrease the traffic issue but will also reduce searching for parking slots. In most of the buildings,
the parking problem faced in immensely crowded places. there are no spaces to be provided to outsiders to park
their respective vehicles. Thus, leading to the
Keywords - Dijkstra’s Algorithm, Geofencing, Genetic
Algorithm, GPS congestion. People also find it difficult to look for a
new parking place or plot. Most of the traffic problems
I. INTRODUCTION occur because of the congestion which arises due to
various vehicles in the urban areas. People waste a lot
This technology provides ways to connect all of time searching. Our system enables the user to get
surrounding environment into network with least the realtime status/availability of parking places much
Human evolvement. Smart Parking System using before setting their journey. Here, the most challenging
navigation is just a start. However, there is a common task is to make the best use of the existing resources
architecture or common solution which every device in such a way that it reduces the time used for
follows it changes according to the way of searching and also traffic congestion issue.
implementation. So according to the context specified,
the android application which will be designed for A few existing parking systems make use of various
Smart parking with GPS has different definitions. Here hardwares like sensors to know the current status about
we are using the concept of Geofencing for the parking slots but it turns out to be expensive, so our
connections between various devices. main aim is to formulate a device in a way which is
more economic, efficient and provides better
These physically connected devices over the network performance.
are called as smart devices. These are smart machines
which communicate and exchange messages in the At some point of time ,we might have observed
form of signals or reminders with other machines, location based alerts pop-up on smart phones while
environment, objects etc. It also incorporates the walking on streets when there is a retailer or a business
connection between two different or similar machines, firm around. Geofencing consists of virtual parameters
between machines and human and vice-versa. This of radial or polygonal shapes. Whenever a user enters
communication is known as M-M communication. As the area, that we have defined, we can track them
M-M communication is developed by the various down. Radial shapes are quite larger than polygonal.
standardised bodies such as Open Mobile Alliance Geofencing enables contextual offerings by providing
(OMA), it equips the daily life things with transceivers, relevant information needed in right time and right
sensors, actuators and micro-controllers etc., so that place. We can draw the Virtual parameters in
these devices could connect to each and interact for Geofencing around any particular area in a map. There
communication. Some points to ponder about internet are many benefits of Geofencing, it increases sales and
of things are: loyalty. It also effects the decisions of purchase by
providing immediate value when needed. Geofencing
1) Tracking Behaviour. is used for promotions, inceptives, survey campaign
Our system focuses on covering various areas
and track campaign. The concept of Geofencing have including covered parking, open parking and street side
been used to track the authenticated users entering into parking.The centralised server is responsible for
a particular area or location. managing the entire smart parking systems information
such as number of slots booked, number of parking
farms, availability of vehicles etc and can be accessed thoughts of natural selection and gene sorts. The
through some secured gateways through network. fundamental concept of Gas is designed to simulate
strategies in a trivial way vital for evolution,
The application provides an interactive and well
particularly those who follow the standards of survival
designed user interface to the client so that clear
of fittest, first laid down by Charles Darwin. As such
information is accessed by the client about the system.
they represent a smart exploitation of a random search
An administrator is responsible for creating new
inside a defined search area to solve a hassle.
parking areas. A person gives the description or
Essentially, numerous random sets of parameters are
information about the parking area
applied to a set of rules and a fitness value
a n d t h e administrator decides whether to add the
(optimisation cost) is calculated for each. Based on this
parking area or not. Only the administer has the power
fitness values, the great units are mixed (this is a
to change the number of parking slots in any particular
combination of selection, Crossover and Mutation)
parking area and even further remove some. pa. The
together and new units are again applied to the set of
updated timing of each parking slot is shown along
rules till the most advantageous parameter(s) are
with unique number. And more importantly, the
obtained.
application also provides the navigation to the client’s
destination.
III. OBJECTIVE
II. OVERVIEW OF ALGORITHMS
The goal of this project is to reduce human efforts by
Dijkstra’s Shortest Path Algorithm identifying the areas where the use of resources is
optimal so that it can provide it’s citizens a better way
The Dijkstra’s algorithm is used to calculate the of living, decent quality life in a clean and sustainable
shortest path between two points using a graph having environment. The ultimate goal is that is to remove the
nodes and edges as its elements. The Dijkstra’s unwanted stress drivers experience as they waste their
algorithm assigns a cost value to each and every node priceless time looking for the optimal parking spot.
of the graph, mostly the value for the initial node is
zero while for other its infinite. Nodes can be divided IV. LITERATURE
into two sets: tentative and permanent. It chooses SURVEY
nodes, makes them tentative,
The algorithm can be presented as: In the literature, there exist many studies that aim for
Steps: smart parking. Every approach uses different
1. Initialise Information Technology concept such as internet of
2. Begin with the source node i.e. the root node of the things, cloud computing, wireless sensor network,
tree. artificial intelligence techniques,
3. It is the first permanent node and hence it is assigned g e o g r a p h i c information systems, android
the value (cost) 0. application. In [1], An algorithm depending upon
4. The neighbour nodes of the last permanent nodes are dispatching rules and varying conditions of parking
thoroughly examined. that guides the smart parking in a city have been
5. Nodes are made tentative, for doing so they are proposed. Chatzigiannakis [2], presents a more secure
assigned a cost value. smart parking application method. It is more trusted and
6. A forest of tentative nodes is created and among fits in place when an untrusted network exists. It also
them; avoids the important information from being
a. Locate the node with the smallest cumulative cost transmitted between various system components.
and mark it permanent. A permanent node is never Another smart car parking system based on cloud
bothered again and its cost recorded now is final. technology using Internet of Things(IoT) has been
b.Select the direction with the shortest cumulative cost proposed in [3]. In the study [4], combination of
only If a node can be reached from more than one UHF-RFID and IEEE 802.15.4
direction. Wireless Sensor Network has been used to present a
6. Until all nodes are made permanent, repeat steps 4 to smart parking system. This system makes the use of
6. RESTful Java and Google cloud messaging concept
7. Finish. has been used along with a mobile application which
provides a way, a user can find a vacant parking
Genetic Algorithm: location.
Samaras [5], has presented a method for approximate
Genetic Algorithms are adaptive heuristic seek calculation of free size of parking places which is
algorithms which can be premised on the evolutionary based on Fuzzy rule. The Fuzzy rules are generated by
using the real data which is collected by the wireless
sensor network. This is an energy efficient method. In
[9], Wang has developed a smart system based on
reservation that allows users to to search and book the
vacant parking slot. The wireless sensor network is
used to get the status of he parking places. This system
emphasises on reducing the search time. The block
diagram for the same is given below.
Fig.1 with the Genetic algorithm which predicts the path
depending upon the history of travel and previously
V. EXISTING SYSTEM booked parking slots.

Down the ages car parking is done the traditional way Along with these two algorithms, the concept of
which includes human intelligence and activity. It is Geofencing has been used. Geofencing takes the
basically performed manually by individuals. This advantage of Global Positioning System (GPS) to
system which includes human activities is not entirely define the boundaries distributed geographically. It is a
safe and sometimes may cause a lot of trouble. The type of alert generation system which enables the
number of vehicles are increasing rapidly which leads
us to the bigger picture; the problem of insufficient
space and no planning on how to park. In a nutshell the
existing system lacks security, and at times may leads
to inconvenience.

V. THE PROPOSED METHODOLOGY

Through this paper we aim to meet the requirements of


a smart city. In general smart city refers to an
environment where human efforts are reduced
considerably almost negligible. Enhancing and
optimising any algorithm or application is the need of
the hour. Efficiently using the resource can make a
difference which reduces human efforts and cost of
working. Mobile applications and mobile computing
for some reasons are becoming common day by day
and has its hold in the technology. It provides a
platform to communicate by computing the real-time
statistic from the physical world. . The block diagram
for the proposed system is given in Fig.2.
The application interface is provided to the client
so that client can access the system. The administrator
can create new parking areas by just providing the
description or information about the parking area. The
administrator can add as well as remove the number of
parking slots in any particular parking area. A unique
number along with the updated timing is shown. The
reason that this system is different is because of the
navigation property. The user reaches to the destination
with the help of a proper navigation.The Dijkstra’s
algorithm is used to calculate the shortest path along
Fig.2 phone will track the location of the user and hence will
navigation will be provided from the user’s location.
administrators to set up triggers each time a device A timer will also be set based on the time estimated
enters or leaves a particular location whose boundaries by the GPS to reach the parking area Dijkstra’s
are defined by the administrator. Mobile devices are algorithm will be used for finding the shortest
actively used to get the position of a client and the distance and genetic algorithm will be used used for
notifications are generated accordingly.[6] suggesting the preferred choice based on previous
booking history of the customer. For the purpose of
VI. APPLICATION counting of cars parked and number of spaces available
WORKING the concept of Geofencing will be used. As soon as
the car enters into the parking area the location of the
In the system which we have proposed, an android customer and parking area will be matched and trigger
application will have two sections named as a user will be activated in which the number of available slots
section and a business section. User section is in parking area will be decreased by one and the check
designed so as to use all the services like getting in time of the customer will be stored on the server and
nearest parking area’s location, booking the slots and timer will be started.At the time the customer takes his
the making payment. Business section will be used car and goes out of the Geofencing are again the
by the owners of the parking area in order to register trigger will be activated based on the mismatching of
their land under the application. Once the the location of the customer and the parking area which
administrator approves the land, it will become the will result in counting of the time for which the car was
part of the parking farms already present. The users parked, calculating the fare based on the basic charges
will use this application not only to stay connected and sending the bill on the customer's screen via the
to the server through internet but also for booking application. The user can pay the bill by number online
the parking slot. The information about the parking methods. Using the other part of application which is
lots will be updated accordingly. The GPS in the smart business login the user can register their available
parking area, free plots or residential parking for the
use of customers at a specified rate also the available
space can be removed or updated by the user.

VII. CONCLUSION

This designed smart parking system is simple,


economic and provides an effective solution to move
towards automation.It provides a smart way to park
vehicles. It can access and map the status of parking
slots from any remote location through web browser
with the help of GPS. Since it can access the location
of any parking slot from any farther place hence,
reducing the risk of looking for the parking slots in
any parking area and also it eliminates unnecessary

Fig.3
travelling of vehicles as navigation is provided. So it
reduces time as well as it is cost effective.

IX. FUTURE WORK

The future scope to adopt this Smart Parking System


(SPS) so that an android application can serve the
purpose of searching the parking lots available at
nearby places. A satellite navigation device can also be
used so that the drivers will always be aware of
whether there are free spaces are not.A way can also be
developed to send some notifications to users on their
respective smart phones when vehicle enters in a
particular location like shopping places and some
streets in a city etc.Dynamic pricing scheme can be
used to calculate the cost for parking.

X. REFERENCES

[1] J.-H. Shin, H.-B. Jun, A study on smart parking


guidance algorithm,Transportation Research Part C,
vol. 44, pp. 299–317, 2014.

[2] I. Chatzigiannakis, A. Vitaletti, A. Pyrgelis, A


privacy-preserving smart parking system using an
IoT elliptic curve based security platform,Computer
Communications, vol. 89–90, pp. 165–177, 2016.

[3] Y. Atif, J. Dinga, M. A. Jeusfeld, Internet of Things


Approach toCloud-Based Smart Car Parking,
Procedia Computer Science, vol. 98,pp. 193-198,
2016.

[4] L. Mainetti, L. Palano, L. Patrono, M. L. Stefanizzi,


R. Vergallo,Integration of RFID and WSN
Technologies in a Smart ParkingSystem, 2014 22nd
International Conference on
Software,Telecommunications and Computer
Networks (SoftCOM), pp. 1-4, 17-19 Sep 2014

[5] I. Samaras, A. Arvanitopoulos, N. Evangeliou, J.


Gialelis, S. Koubias,A fuzzy rule-based and energy-
efficient method for estimating the free size of
parking places in smart cities by using wireless
sensor networks,2014 IEEE Emerging Technology
and Factory Automation (ETFA), pp.1-5, 16-19 Sep
2014.

[6] Faheem1, S.A. Mahmud, G.M. Khan, M. Rahman


and H. Zafar,ǁ‖A Survey of Intelligent Car Parking
Systemǁ‖, October 2013

[7] Peter G. Anderson, Introduction to Genetic


Algorithms, Rochester,New York,
2002

[8] Sam Hsiung, James Matthews, “An Introduction to


GeneticAlgorithms,”

[9] Sachith Abeysundara, Proceedings of the


International Conference on Information and
Automation, December 15-18, 2005, Colombo, Sri
Lanka “A Genetic Algorithm Approach to Solve
theShortest Path Problem forRoad Maps”

[10] whatis.techtarget.com/definition/geofencing4

You might also like