Content uploaded by Priyankar Roychowdhury
Author content
All content in this area was uploaded by Priyankar Roychowdhury on Jan 17, 2016
Content may be subject to copyright.
II2202, Fall 2015, Period 1-2 2016-01-18
1
Smart Urban Traffic Management System
Sarjo Das
KTH Royal Institute of Technology
Stockholm, Sweden
sarjod@kth.se
Priyankar Roychowdhury
Independent Research Scholar
Kolkata, India
priyankarroychowdhury3@gmail.com
Abstract— In recent years the ownership of private vehicles has
increased many folds, which is causing difficulty in management
of urban traffic. Traffic management is the focus area for most
urban dwellers and planners. Some of the main concerns for
traffic management of big cities is traffic congestion and
avoidance, as these issues cause huge damages on both personal
and environmental level. Moreover, in many cities traffic signal
lights at crossing points are timer based which is inefficient
method of controlling traffic. This paper presents a more efficient
approach of managing urban traffic with the help of intelligent
traffic management system which uses internet of things to
achieve this. This method of intelligent traffic management uses
components like RFID and detection technologies to sense the
presence and movement of tagged objects, the traffic will be
monitored and managed automatically using this system. The
data collected from this system will be sent to a centralized
system for further analysis. Moreover the traffic signal lights at
crossing points are based on traffic density of roads intersecting
at that point. The paper proposes an architecture which
integrates Internet of things and other moving components like
data management techniques to create a model for traffic
management and monitoring. The model will constitute of a
single platform where this platform will communicate with the
large number of decentralized heterogeneous components.
Keywords: RFID, Single Board Computer, IOT, Cloud, Intelligent
Transport, Road management, Road planning, City planning
I. Introduction
Smart City [1] is a term which is associated with various
innovative technologies in order to make a city ―smarter‖ to
improve the quality of life of people. Among the different
dimensions that make a city smart, one of the very important
one is transportation. Smart Transportation deals with creation
and implementation of an intelligent traffic management
system that deals with congestion detection and avoidance,
emergency management, car safety and accident prevention
etc. It also tries to make transportation greener by helping to
reduce gas emissions, fuel or energy consumption in vehicles.
Transport sector contributes to 27% of energy related to CO2
emissions and is the fastest growing sector in terms of
greenhouse emissions in developing countries [2].
Fig.1. Number of cars sold worldwide from 1990 to 2015
(Source:
http://www.statista.com/statistics/200002/international-car-
sales-since-1990/
on 13 Jan, 2015)
As seen from Fig.1. the number of cars being sold continues to
increase around the world. This gradual increase in the
number of cars sold has already put a huge strain in the
infrastructure of current roads. Moreover, the increase in total
road area is much slower compared to the number of cars
being purchased and registered. The problem of traffic
congestion and long waiting queues are quite common along
with the significant rise in the number of mishaps and traffic
accidents.
There is an urgent requirement of a Smart System for traffic
management by employing smart monitoring of road systems
that will automatically manage the traffic signalling system in
order to reduce congestion to the minimum taking into account
II2202, Fall 2015, Period 1-2 2016-01-18
2
the traffic density, volume and informing drivers about which
route is best at any given time of the day. Moreover,
government in different countries like India are planning smart
cities and smart transportation system is one of the most
important part of any smart city [3].
Fig. 2. Explains the economic impact due to Congestion which
is more than trillions of dollars for Europe and US combined
in time period of 2013-2030.
Fig.2. Economic impact due to road congestion in Europe and
USA: 2013-2030 (Source:
http://inrix.com/economic-environment-cost-congestion/ on 1
Dec, 2015)
This paper consists of the following sections: ―Literature
Study‖ which describes the related studies, ―Problem
Statement‖ section illustrates the problem we address in this
paper. ―Methodology‖ section portrays the methods and
approach followed to develop the proposed model,
―Framework and Design‖ section explains the functional block
diagram of the proposed model. ―Result and Analysis‖ section
contains our proposed Smart Road Management System
Model and its advantages. The ―Discussion‖ section contains
the sustainability and ethics related to our model and its
economic value.
II. LITERATURE STUDY
A lot of researches have come up with the solutions to
implement Smart Road Traffic Management system and as a
result several approaches have been developed.
Zeng, et Al. [4] discussed the design, framework and
functional modules of an intelligent transport management
system in important traffic hubs inside of a city or for external
transport by means of integration of multiple technologies in
order to automate decision making and increase the utilization,
safety and comfort level. On the other hand, Al-Sakran [5]
proposes a low cost highly scalable and compatible intelligent
traffic administration system framework that is based on
internet of things. The framework uses RFID, wireless sensors
technology, cloud computing, GPS and other advanced
technologies to implement the intelligent traffic management
system. To reduce traffic congestion, road accidents and
increase traffic productivity and efficiency, Bitam, et Al. [6]
explains a cloud computing model that can be added to
intelligent transportation system to improve traffic
management.
To study efficiency of moving vehicle detection by RFID,
Lee, et Al. [7] investigates about RFID read latency and thus
the effectiveness of on-vehicles reader installations for a wide
range of speeds. The impact of readers and tags‘ relative
positions on read errors and read rates are studied
experimentally. The road experiments are conducted at
varying speeds. The results reveal the influence of critical
factors on on-vehicle RFID read performance, and provides
guidance to identify and pursue directions for improvement.
From commutator point of view, for finding out the most
appropriate roads for the motorists, Mbodila, et Al. [8]
discussed in their paper about the implementation of a model
for vehicle traffic monitoring between Johannesburg and
Pretoria in South Africa. Wireless Sensor Network is used as a
tool to to control traffic signals along the roadsides while the
RFID scanner is used to identify vehicles in the congestion
areas so that the traffic officer at the Traffic Monitoring
Centre (TMC) can make use of Global Positioning System
(GPS) by establishing the position of the vehicle at any point
on the roads and determine appropriate road for motorists.
From the big picture view-point, Djahel, et Al. [9] in their
publication presents an up-to-date review of the different
technologies used in the different phases involved in a Traffic
Management System and discuss the potential use of smart
cars and social media to enable fast and more accurate traffic
congestion detection and mitigation. The security threats that
can jeopardize a traffic management operation are also
discussed here. Some open challenges are discussed and their
II2202, Fall 2015, Period 1-2 2016-01-18
3
own vision is also stated to develop robust Traffic
Management System for future smart cities.
III. Problem Statement
The fact which encouraged us to conduct this research is that
in many cities of the world, signal allocation is still based on
timer. The timer approach has a drawback that even when
there is less traffic on a road, green signal is still allocated to
the road till its timer value falls to 0 while traffic on another
road which is more, faces red signal at that time which causes
congestion and time loss to commutators. Most of the present
systems are not automated and are prone to human errors. The
main objective of this paper is to create a better road network
system within the city for smoother transition of traffic to
increase the overall productivity of a city.
IV. Methodology
In this paper, a method has been proposed in which the traffic
signal light controlling decision is performed based on the
traffic density of the roads which intersect each other at a
crossing point. We are influenced to propose this new model
of traffic management system from the works of
Roychowdhury P. and Das S. which is published as
‗Automatic Road Traffic Management System in a City‘,
‗Trends in Transport Engineering and Applications‘, Vol 1,
Issue 2.
To track a vehicle, a sensor is required and in our work to
identify a specific vehicle, a vehicle ID is required. RFID can
effectively meet our requirements and the work of
Viswanathan. B. and Sukhadha. V. published as ‗Intelligent
Traffic Control and Vehicle Tracking System Using RFID‘ in
‗International Journal of Electronics Communication and
Computer Engineering‘ Volume 4, Issue 5 has encouraged our
decision to use RFID for tracking vehicles. While the current
trend is moving towards Internet of Things and Cloud
Computing, Al-sakran‗s work published as ‗Intelligent Traffic
Information System Based on Integration of Internet of Things
and Agent Technology‘ in ‗International Journal of Advanced
Computer Science and Applications‘, Vol. 6, No. 2 has
influenced us to use Cloud Database and RFID and Single
Board Computer (SBC) [10] interfacing to Cloud to propose
our model based on traffic management by Internet of Things.
After going through these publications, it was decided that we
should take the qualitative approach because our proposed
model is based on taking the best features of some of the
existing models and additional features from our own
research. The model proposed in this paper is purely
theoretical and we hope to run some simulations using this
model in the future to get quantitative data. However, for this
paper we decided to follow the qualitative approach due to the
availability of limited time and resources.
V. Framework and Design
The framework and design of our proposed system is as
follows:
The RFID readers acquire the data and send it to the Cloud
[11]. In the place of Crossing Server, single board computer is
used which is also connected to the cloud. Instead of data-base
of each road stored in the corresponding Crossing Server, in
this work Cloud Database [12] is used. Each light of a traffic
signal is connected to an output port of the corresponding
single board computer and no micro-controller needs to be
used.
As proposed the model has various components in conjugation
with internet of things. The model has a core system which
constitutes of various parts. Below is the figure which shows
the component involved in this model.
Fig. 3. Functional Block Diagram of our system
II2202, Fall 2015, Period 1-2 2016-01-18
4
A. Data Sources: Data gathering will be done via the
help of RFID sensors, these sensors will be
embedded in the roof of the vehicles. Data gathering
interrogators will be placed on the entry and exit
points of a path way. These entry and exit points will
serve as the data sources for the system.
B. Cloud: The data is sent to the cloud that
is responsible for processing the data and taking the
major decisions in regards to determining if the road
is congested or there are some other issues with the
traffic flow.
C. Data aggregation and processing: Data aggregation
and processing is connected closely with data
storage. Before storage of data it will be processed
within the cloud.
D. Data storage: Data gathered from data sources will
then be moved to a cloud database after data
processing. This data can be used on demand for
further analysis and planning of future roads.
E. Data analysis: This component is used for analysis
of existing data present on the system. Relevant data
is picked up for analysis and fed it to data analytical
component of the model for analysis. Analysis can
help plan a better system for the existing
infrastructure or help us in designing new
infrastructure.
F. Single Board Computer: It will be used to fetch real
time traffic information from the cloud for
controlling the traffic lights.
G. Consumers: The end product after processing of data
is delivered to the customer. The consumer can get
on demand access to real time traffic data. The
consumer can be the Traffic Control Room or any
commuters who wishes to see the traffic conditions in
his/her smartphone.
VI. Result and Analysis
A. Our Proposed Model
At the entry and exit of a road and at entry and exit of a
parking lot, fixed RFID readers [13] are placed to detect
movement of vehicle at the respective points. The readers
which are placed at entry and exit of a road are called
Crossing Readers (CRs) [14] and which are placed at the entry
and exit of parking lots are called Parking Readers (PRs). CRs
are of two types: a) Exit CRs which are placed at the positions
of road from where vehicles are exiting out of the road b)
Entry CRs which are placed at the positions of road where
vehicles are entering into the road. In case of parking lots,
there are two possibilities: (i) entry and exit points can be
same, (ii) entry and exit points can be different. For case (i), a
single type of PR has to be used to detect entry and exit of a
vehicle which is cat-1 PR while for case (ii), there will be two
PRs: entry PR at the entrance of parking lot and exit PR at the
exit point of a parking lot which constitutes cat-2 PR.
In our model, active RFID tag is placed on the roof of the
vehicle at 0 degree to the horizontal [7]. This tag is coded with
registration number of the vehicle which is then read by the
RFID Readers as the vehicles approach the reader points [15].
On the other hand, the RFID readers has to be placed above
the road at some height greater than the maximum possible
height of a vehicle in that country. Patch angle of RFID
Reader antennas=70 degree from horizontal [7] because area
that can be read will be maximum at this angle.
Fig. 4. Sample RFID Reader
(Source:
http://ops.fhwa.dot.gov/publications/fhwahop12016/sys_design.htm
on 1 Jan, 2016)
Data is gathered via RFID tags, each vehicle has a far field tag
embedded in it. As soon as the tag crosses an interrogator it
transfers the data on it to the interrogator. RFID tag has a
transponder which wakes up as soon as it gets in contact with
the transceiver present in the interrogator as transceiver
provides it with necessary power to work with [16]. This point
of time the tag supplies interrogator with all the required data
of the entity it is embedded with. The data which is present
with the RFID interrogator is sent to the cloud based servers
using the internet. Each node within an area connects to a
wireless router which acts as a natting device connecting a
cluster of RFID‘s to a server. The data is processed and stored
in a readable format. Data Stored on the cloud system is
processed on the cloud and supplied to a single board
computer using routing via Internet. Single board computer
has direct connection to the traffic signal, which in the end
supplies traffic lights with the required information.
II2202, Fall 2015, Period 1-2 2016-01-18
5
Fig. 5. A sample model of an urban four point road crossing
The above diagram represents sample portion of a city road
where there is a crossing point. Entry type CRs are placed at
the entry positions of roads which are 2, 4, 6, 8, 22, 24, 18, 20.
Exit type CRs are placed at exit positions of roads which are 1,
3, 5, 7, 21, 23, 17, 19. The white boxes with the letter ‗P‘
written inside represent parking lots. PRs are placed at
positions 9, 10, 11, 12, 13, 14, 15, 16. At position 14, the PR is
entry type PR while in position 15, the PR is exit type as they
are entry and exit points of a single parking lot respectively.
PRs at positions 14 and 15 constitute cat-2 PRs while the
remaining PRs are cat-1 PRs.
[17] Each CR and PR are assigned to a definite road. In this
diagram, the following are the associations of CRs and PRs
with the roads:
a) road A: 3, 4, 13
b) road B: 1, 2, 9, 10
c) road C: 5, 6, 11, 12
d) road D: 7, 8, 14, 15, 16
e) road E: 17, 18
f) road F: 21, 22
g) road G: 19, 20
h) road H: 23, 24
In the Cloud, for each road, there is a specific database.
[17] Whenever a vehicle enters a road, the entry type CR
sends its ID to the road database and the Traffic Density of the
road is increased by 1. Whenever a vehicle exits a road, the
exit type CR sends its ID to the road database. In the road
database, the ID of the vehicle which already exists being
received from entry type CR is deleted on receiving the same
ID from exit type CR. The traffic density is then decreased by
one.
In case of cat-1 PR, the PR detects a vehicle and sends its ID
to the corresponding road database.
a) If the ID is present in database of the road to which the PR
is associated, it implies that the vehicle is exiting from road
towards parking areas. Its ID is deleted from the road database
and Traffic density of the road is decreased by one.
b) If the ID is not present in database of the road to which the
PR is associated, it implies that the vehicle is entering road
from parking areas. Its ID is added in the road database and
traffic density of the road is increased by one.
In case of cat-2 PR, when an entry type PR detects a vehicle, it
indicates a vehicle is leaving the road and entering the parking
lot. The PR sends the ID of vehicle to the road database. This
ID is already present in the database of the road, since it was
added to the database of the road when the vehicle entered the
road. Now, on receiving the same ID from entry type PR, it is
deleted from the road database and the road density is
decreased by 1. When an exit type PR detects a vehicle, it
indicates a vehicle is entering the road from the parking lot
and its ID is added to the road database and the road density is
increased by 1.
1. Data-analysis, processing and aggregation
In a crossing, signals have to be provided to roads with Exit
CRs to control the vehicles that exits from these roads and
enters roads which have Entry CRs at the crossing.
The following data operations are performed at the beginning
of signalling cycle at the crossing point:
a. Traffic Density of each road with exit CRs is checked
and arranged in descending order.
b. Green signal is allocated to the road which is first in
the order till its Traffic Density decreases by a certain value or
reaches 0 whichever is earlier (This certain value is the
number of vehicles from the road that are allowed to pass
through the crossing at a single signalling cycle).
c. Repeat step no. b for the remaining roads placed in
the order.
d. Repeat step a.
2. Detection of a Vehicle Violating Signal
[18] When the signal of a road with entry CR at the crossing
point is Red, if it (entry CR) detects any vehicle at that time,
its ID is sent to the Vehicle Violation Database contained in
the Cloud Database.
II2202, Fall 2015, Period 1-2 2016-01-18
6
3. Detection of an anomaly
Many things have been considered to analyse the viability of
this model, for example we have integrated timer values for
the counters. If a car stays for too long between entry and exit
CR and don‘t enter any parking PR, administrators of the
system will be alerted by the core system.
B. Analysis
Our model provides an advantage over the model in the paper
‗Intelligent Transport Management System for Urban Traffic
Hubs Based on an Integration of Multiple technologies‘ [4] in
the sense that we have proposed a model which is versatile
and can be implemented in any city and country around the
world. Also, it has an automatic traffic management system
which requires no monitoring but the traffic condition can be
tracked by anyone, commutator or authority. While Zeng et al.
[4] in their paper has proposed a system which is based only
on traffic hubs inside a city. It also uses a number of
technologies which will increase cost to a great extent while
our model only uses RFID Readers and Tags on vehicles and
Single Board Computers at each crossing point of roads which
are not an expensive infrastructure. Also Cloud Computing is
cost-effective than any existing connected computing
technology because Cloud can be leased or hired from Cloud
Service Providers and no new infrastructure for this purpose is
even required.
Al-Sakran [5] in the paper ‗Intelligent Traffic Information
System Based on Integration of Internet of Things and Agent
Technology‘ has proposed an intelligent system for traffic
monitoring while our proposed system is not only for traffic
monitoring but it is mainly for traffic controlling of the whole
city.
Mbodila, et Al [8]. In their publication ‗Implementation of
novel vehicles' traffic monitoring using wireless sensor
network in South Africa‘ have described about the
implementation of a model for vehicle traffic monitoring
between Johannesburg and Pretoria in South Africa. This
system can find out appropriate road for motorists. However,
in our model we are not doing any research about appropriate
road for motorists because Google Maps [19] already exist
which in navigation mode can do the same function for a
motorist and we have decided not to research something that
already exists.
The framework explained here is in preparatory stage. There
are numerous degree for changes. The framework must be
adjusted to meet the logistical necessities which contrast from
street to street and city to city. A testing environment can be
developed under ideal conditions for testing and development
of the model as well as addition of a manual override.
VII. Discussion
A. Sustainability and Ethics
The model which we are implementing provides an optimized
traffic system which in the end delivers a transportation
system producing lesser CO2 emissions. The road transport
vehicles will have better infrastructure to use and in the end it
will result in lesser driven kilometres [20]. The sustainability
aspect of this model is one of the key features associated to it
as this System also helps in reduction of time consumed for
travel. Traffic management system proposed in our model
system enable us in planning of future road infrastructure and
helps us to create a sustainable ecosystem.
The data gathering and monitoring system will be designed in
such a way that it will not have any conflict of interests with
an individual i.e. data gathering RFID will not contain any
information about an individual, it will just contain
information in regards to the identity of the vehicle not the
owner. Authorities can correlate this information with their
own system in order to attach an owner's identity to a vehicle.
The data collection servers will be located in a secure location
in order to maintain data integrity.
B. Economic Value
A critical part of developing a smart city is Traffic
management. Traffic management and road planning are very
much required by the city planners to keep the city free from
traffic congestion and also to make the road infrastructure
future proof to accommodate increasing number of vehicles.
The relevant model discussed above can play a vital role in
traffic management. The system can be deployed at various
road crossings of the smart city with interconnected central
server system. The conceived system will yield greater return
in the long run both in terms of ROI (Return on Investment)
and TCO (Total Cost of Ownership) because of the following
reasons:
1. Number of deployed persons required to manage traffic is
much less. As the system is fully automated, only a handful of
people can manage the entire traffic of the city.
2. Less waiting time and reduction in traffic congestion will
help people to save fuel and will reduce pollution.
3. There is no room for human errors, so the number of road
accidents will be reduced considerably.
VIII. Conclusion
The paper gives a real time traffic monitoring and
management model to solve the problem of traffic congestion
in urban areas. The proposed model uses technologies like
IOT, RFID, cloud computing and other advanced technologies
to collect and analyse real time traffic data.
II2202, Fall 2015, Period 1-2 2016-01-18
7
The proposed system can help reduce traffic jams and waiting
times at traffic lights and achieve traffic fluency. By entirely
automating the control of traffic lights by relying on the
intelligent traffic management system human intervention can
be brought down to minimum. The traffic data collected and
analysed can be used by the traffic control rooms to make
controlling traffic easier or it can be used by city developers to
plan and develop future roads and cities.
Intelligent traffic management system has already been
implemented in cities like New York [21]. There is huge scope
in the field of intelligent traffic management system. The
model explained in this paper also explains a way in which an
Intelligent and Smart Traffic Management System can be
implemented.
IX. References
1. DIGITAL AGENDA FOR EUROPE-Smart Cities
https://ec.europa.eu/digital-agenda/en/smart-cities (Last
accessed on 1st December, 2015)
2. Bridging the Gap www.transport2012.org (Last accessed on
1st December, 2015)
3. SMART CITIES MISSION- Ministry of Urban
Development Government of India http://smartcities.gov.in
(Last accessed on 1st January, 2016)
4. Zeng, N., Qin, K. J., & Li, J. (2010, October). Intelligent
transport management system for urban traffic hubs based on
an integration of multiple technologies. In Industrial
Engineering and Engineering Management (IE&EM), 2010
IEEE 17Th International Conference on (pp. 1178-1183).
IEEE.
5. Al-Sakran, H., (2015). Intelligent Traffic Information
System Based on Integration of Internet of Things and Agent
Technology. International Journal of Advanced Computer
Science and Applications (IJACSA), Vol. 6, No. 2, 2015 (pp.
37-43).
6. Bitam, S., & Mellouk, A. (2012, December). Its-cloud:
Cloud computing for intelligent transportation system. In
Global Communications Conference (GLOBECOM), 2012
IEEE (pp. 2054-2059). IEEE.
7. Lee, E. K., Yoo, Y. M., Park, C. G., Kim, M., & Gerla, M.
(2009, September). Installation and evaluation of RFID
readers on moving vehicles. In Proceedings of the sixth ACM
international workshop on VehiculAr InterNETworking (pp.
99-108). ACM.
8. Mbodila, M., Obeten, E., & Bassey, I. (2015, June).
Implementation of novel vehicles' traffic monitoring using
wireless sensor network in South Africa. InCommunication
Software and Networks (ICCSN), 2015 IEEE International
Conference on (pp. 282-286). IEEE.
9. Djahel, S., Doolan, R., Muntean, G. M., & Murphy, J.
(2015). A Communications-oriented Perspective on Traffic
Management Systems for Smart Cities: Challenges and
Innovative Approaches. Communications Surveys &
Tutorials, IEEE, 17(1), 125-151.
10. Hässig, S. Single-Board Computer (SBC) Hardware and
Software Capabilities for Real-Time Processing.
11. What is cloud computing?
http://www.ibm.com/cloud-computing/what-is-cloud-
computing.html (Last accessed on 1st December, 2015)
12. Arora, I., & Gupta, A. (2012). Cloud databases: a
paradigm shift in databases. International J. of Computer
Science Issues, 9(4), 77-83.
13. The Basics of an RFID system, atlasRFIDstore, pp10.
http://rfid.atlasrfidstore.com/hs-fs/hub/300870/file-
252314647-pdf/Content/basics-of-an-rfid-system-
atlasrfidstore.pdf last accessed on 1st December, 2015)
14. Roychowdhury, P., & Das, S. (2014). Automatic Road
Traffic Management System in a City. Trends in Transport
Engineering and Applications, 1(2), pp. 38
15. Measuring Border Delay and Crossing Times at the U.S.–
Mexico Border—Part II
http://ops.fhwa.dot.gov/publications/fhwahop12016/sys_desig
n.htm (Last accessed on 1st January, 2016)
16. How Does RFID Technology Work? [Technology
Explained]http://www.makeuseof.com/tag/technology-
explained-how-do-rfid-tags-work/ ( Last accessed on 1st
December, 2015)
17. Roychowdhury, P., & Das, S. (2014). Automatic Road
Traffic Management System in a City. Trends in Transport
Engineering and Applications, 1(2), pp.39
18. Roychowdhury, P., & Das, S. (2014). Automatic Road
Traffic Management System in a City. Trends in Transport
Engineering and Applications, 1(2), pp. 40
19. What is Google Maps?
https://manuals.matrix.squizsuite.net/google-
maps/chapters/what-is-google-maps (Last accessed 16th
December, 2015)
20. Reducing CO2 emissions from cars and vans
https://www.acea.be/uploads/publications/06052015_ACEA_
Tackling_Road_Transport_CO2_Emissions_paper_cm.pdf
(Last accessed on 1st December, 2015)
