Rh Research

A More Efficient Way to
Control Traffic Lights

Through AI-Led Smart
City Management
-R.H.DHRUVA
1JB22CS114
INTRODUCTION
 More automobiles on the road and growing urbanization have
increased air pollution, travel times, fuel consumption, and
traffic jams. In this situation, integrating artificial intelligence
technology into traffic management systems is a viable
approach to effectively and efficiently resolving these issues
 Using the power of artificial intelligence algorithms and data
insight, traffic lights can be empowered to dynamically adjust
their timing model according to prevailing traffic conditions [5].
Such adaptability can significantly reduce congestion, minimize
travel time and ultimately contribute to a greener and more
sustainable urban environment .
 Different artificial intelligence techniques that can be used for
real-time traffic analysis and decision-making are analyzed,
such as machine learning models, computer vision, and sensor
fusion [7]
RELATED WORK
 Cameras mounted on traffic signals or poles along the side
of the road are used by computer vision-based systems to
capture real-time traffic images, which are then processed
by AI algorithms
 A study by researchers at the University of South Florida
revealed a 50% reduction in intersection delay by
implementing a computer vision based system
 the University of Texas implemented an AI-based system
that reduced travel time by 25% and fuel consumption by
18%.
 This study compares the efficacy of hard code and
dynamic coded methodologies, with the dynamic algorithm
exhibiting a 35% improvement [2]
PROBLEM STATEMENT
 The challenge of traffic congestion and the less-than ideal
flow of vehicles is what AI-controlled traffic lights are meant
to address.
 Road congestion is recognized to result in several
drawbacks, including longer travel times, lower
productivity, greater fuel consumption and emissions, lower
safety standards, and more wear and tear on infrastructure.
 Traditional traffic signal systems have frequently relied on
set timing or pre-existing schedules that are inefficient
because they cannot adjust to changing traffic
circumstances in real time. Real-time traffic data can be
analyzed through the application of artificial intelligence
(AI) in traffic light control, allowing for the corresponding
adjustment of traffic signal timings and patterns.
PROPOSED METHOD
Three-tier software architecture, sometimes called three-tier architecture, was developed in the 1990s
to address the shortcomings of two-tier design.Unlike the two-tier design, which supports only 100
users, the middle tier can serve hundreds
A three-tier architecture is used when a distributed client/server design is required, which effectively
improves speed, flexibility, maintainability, reusability, and scalability by hiding the complexity of
distributed processing from the user
A use case diagram, a form of UML diagram, is a tool that shows how different things, such as actors
interact with a particular system.
IMPLEMENTATION AND RESULT
 The training dataset was created by manually labeling images scraped from
Google using the graphical image labeler LabelIMG [12]
 The output neurons in the final layer were chosen exactly to match the number of
classes the model is expected to distinguish, in this case, four: car, bicycle,
bus/truck, and rickshaw. In addition, the number of filters was increased using
the formula 5* (number of classes of 5), resulting in a total of 45
 You Only Look Once (YOLO), which provides the necessary processing speed and
accuracy for vehicle recognition, is used
 The following test has been tested by using the proposed methodology.
1.To ensure the machine learning algorithm used to optimize signal timings
is functioning correctly.
2.To ensure the real-time traffic data is correctly integrated into the machine
learning algorithm.
3.To ensure that the system is reducing travel times for vehicles passing
through the intersection.
4.To ensure that the system is improving fuel efficiency for vehicles passing
through the intersection.
5.To ensure that the system is reducing emissions at the intersection.
CONCLUSION
The suggested AI-based traffic signal control system, in
summary, has enormous potential for enhancing urban
traffic flow and easing congestion. This system can
adaptively regulate traffic signals by utilizing real-time data
and sophisticated machine learning algorithms, resulting in
more effective transportation and improved quality of life
for city dwellers.

Rh Research

Uploaded by

Copyright:

Available Formats

Rh Research

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rh Research

Uploaded by

Copyright:

Available Formats

A More Efficient Way to

Control Traffic Lights

You might also like