DYNAMIC TRAFFIC CONTROL

SYSTEM

PROJECT SYNOPSIS
OF MAJOR PROJECT

BACHELOR OF TECHNOLOGY
Computer Science and Engineering

SUBMITTED BY: GUIDEDE BY:

Sandeep Kumar (201195) (2001333) Prof. Bhavneet Kaur Chahal

Saurabh Suman (201254) (2001339)
Saurav Vinod Sharma (201058) (2001340)
Abhiraj Kumar (201269) (2028511)

Estd. By SGPC
In 1993

Baba Banda Singh Bahadur Engineering College Fatehgarh

Sahib
September, 2023
Table of Content: Pg.no.

1. Introduction:………………………………………………………………… 2-3
1.1. Title
1.2. Traffic Light System

2. Literature Survey:………………………………………………………….. 4-5

3. Methodology/ Planning of Working: ……………………………………... 6-6

4. Facilities required for proposed work: …………………………………… 7-8

4.1. Facilities
4.2. Software
4.3. Hardware

5. Reference for this Project:…………………………………………………... 9-10

5.1. Traffic Signal Control Optimization
5.2. SCAT
5.3. ATCS
5.4. SCOOT
5.5. Genetic Algorithm in TSO
5.6. Deep Neural Network in TSO

1|Page
Introduction:

Title: Dynamic Traffic Light Controller System

Traffic congestion is a serious problem in many cities and rural areas also. So, we can say that around the
world. Traffic congestion is a major challenge in the many and most populated cities. When we travel to two
different places within the same city then it is difficult for the traveler in traffic. Sometimes due to traffic
congestion problems the people lose many things like time, opportunities. Traffic congestion also directly
impacts the industries areas. Due to traffic problem, there is a loss of productivity from workers are effect by
traffic congestion, delivery of products is also affected by traffic because delivery gets delayed. All the above
thereby the costs go on increasing.
To solve the congestion problems, new facilities have to been developed and provide new infrastructure also,
but at the time it makes difficult again. When we can provide new infrastructure there is a big disadvantage of
making new roads is that the surroundings become more congested. That’s why we must improve the system
instead of making new infrastructures. For example, most of the countries are working on their existing
systems to resolve these problems. When we can manage transportation then mobility also improves the safety
and traffic flows. When we can enhance route guidance systems, public transport, traffic signal improvements
and incident management, congestion problem can be reduced from the analysis of US department of
transportation, it has been found that one major reason of congestion is the reoccurring of congestion. Due to
reoccurring congestion, the roads have been used repeatedly. The non-recurring congestions which are caused
due to traffic incidents, special events, work zones, weather etc. Reoccurring events reduces the capacity and
reliability of the transportation system.
The main goal of traffic research is to optimize traffic flow of people and goods also. As the number of road
user’s increases and resources of infrastructures are limited, the intelligent control of traffic will be a very
serious issue in the future. There are several models for traffic simulation. Traffic light optimization is a
complex problem. With multiple junctions, the problem becomes more complex. The flow of traffic constantly
changes which depends on the time of day, the week, and the year is also a complication. Roadwork and
accidents further influence complexity and performance. Our main concern is with the Traffic Control Light
System. The Traffic Lights are explained as below.

Traffic Light System.

A traffic light system is used to minimize the traffic on road. Traffic safety equipment are used to road safety
like as signalling device positioned at an intersection and road divider or other location in order to indicate
when the traveller must ride, drive, walk. The traffic lights commonly have three main light colours, such as
the red light is stop and the meaning of green light is go and yellow that means ready to stop. For the
pedestrians, there are only two light colours one is red and other is green light. There are many benefits of
traffic light system, besides reducing the number of accidents. The government makes few rules to overcome
this problem such as punished to all the traffic offenders. The traffic lights placed at the location where risk of
accident is high or large jam is creating. Increasing the numbers of traffic lights also have problems those are
as below:

a) The heavy traffic jams

With advance in technology, the number of vehicles on roads are increased, those creates the traffic jams. This
happened usually at the main intersections in the morning time, before and after the office hour. So that’s all
reasons are affect the time of people.

2|Page
b) The road user waits
The traffic light used as road safety equipment on road, but sometimes traffic lights is the cause of wasting of
time for people. At the certain junction, but sometimes the red light is occur but there is no traffic, the road
users should wait until the red light convert into green light. If peoples run the red light, then they have to pay
the fine.

c) Emergency vehicle stuck in jam

Because of traffic jams, the emergency vehicle, such as police cars, ambulance and the fire brigade stuck at the
traffic lights, because the users are waiting for green traffic light. It is very critical problem, emergency
become more complicated.
As above, we mentioned the main problems with the traffic lights. There are various techniques to solve these
problems. All these techniques are reviewed under the literature survey.

3|Page
Literature Survey:
Leena Singh et al. in [1] Real-time traffic signal control is an integral part of the urban traffic
control system and providing effective real-time traffic signal control for a large complex traffic
network is an extremely challenging distributed control problem. The model is developed using
genetic algorithm implemented in MATLAB. The model developed is based on the genetic
algorithm, which optimizes traffic signal timings in real time and provides a set of optimum green
time extension for all the four phases depending upon the surveyed traffic conditions. an “intelligent”
isolated intersection control system was developed. The developed intelligent” system makes “real
time” decisions as to whether to extend (and how much) current green time.

Javier J. Sanchez et al. in [2] The model is based on three basic design items: The use of Genetic
Algorithms as an optimization technique, the use of Cellular Automata Simulators within the
Evaluation Function, and the use of a Beowulf Cluster as parallel execution environment for this
architecture Up to date traffic network optimization has been faced using the Trial-And-Error
method. This method cannot ensure that the whole searching space is covered. We propose a new
method, a non-deterministic optimization one to solve this task. In this work we optimize traffic light
cycles.
Halim Ceylan et al. in [3] The genetic algorithm approach to solve traffic signal control and traffic
assignment problem is used to tackle the optimization of signal timings with stochastic user
equilibrium link flows. Signal timing is defined by the common network cycle time and the offsets
between the junctions. The objective function adopted is the network performance index (PI) and its
use for the Genetic Algorithm (GA) is the inversion of the network PI, called the fitness function.
The computation results show that the GA approach is efficient and much simpler than previous
heuristic algorithm. Furthermore, results from the test road network have shown that the values of the
performance index were significantly improved relative to the MC.

Javier J. Sánchez-Medina et al. in [4] optimization techniques are used based on the combination
of three key techniques: 1) genetic algorithms (GAs) for the optimization task; 2) cellular-automata-
based micro simulators for evaluating every possible solution for traffic light programming times;
and 3)Multiple-instruction–multiple-data(MIMD) also called beowulf cluster multicomputer of
excellent price/performance ratio described positive experience with the optimization of traffic light
cycles in a real city area. A distinctive feature of this paper is the large scale of the underlying grid.
Using the supplied maps and statistics, we have simulated the present-day traffic behaviour.
Additionally, we have optimized the traffic signal times, yielding better results with regard to several
predefined parameters.

Suhail M. Odeh et al. in [5] an intelligent traffic light that controls the flow of traffic was
introduced. System detects the level of congestion and the abnormal situations in two main highways

4|Page
and four intersections. Collects the data and the information from a video imaging system by system,
it is used to captures and interprets images, this image is used to detect and count the vehicle. This
data will be sent to another system based on genetic algorithm. System used in the smart city
platforms to manage the transportation system through the control of the intelligent traffic lights. The
system depends on some theories and rules that made to order the priority and the green light interval
time, genetic algorithm used in this proposed system and input to GA is based on the data that got
from the video image detection
Martin Kelly et al. in [6] the values of the parameters governing the simulations are identified
through the use of a genetic algorithm. In a first instance, we will continue pursuing experiments with
the current model in order to identify additional optimums; then we will extend the current
experiments by comparing with other techniques and evaluating the current model under different
congestion situations. Second, another model is planned which will tackle re-routing of emergency
vehicles only. We are also planning to combine these two models together where both regular and
emergency vehicles are re-routed and traffic globally optimized. Third, as most simulations of car
traffic control, we are using a square grid of routes for modelling the city.

5|Page
Methodology/Planning of Work:
The project will follow these key steps:

a. Data Collection and Integration:

Set up data collection points using cameras, sensors, and GPS devices.
Develop data integration algorithms to process real-time traffic data.

b. Machine Learning Models:

Design machine learning models to analyse traffic patterns and predict congestion.
Develop algorithms for dynamic signal timing adjustments.

c. Eco-Friendly Optimization:
Incorporate environmental parameters in signal timing optimization algorithms.
Monitor and measure carbon emissions reduction and fuel savings.

d. User Interface:
Create a user-friendly interface for traffic management personnel.
Implement manual control features for intervention when necessary.

e. Testing and Validation:

Deploy the system in a controlled test environment.
Collect performance data and fine-tune algorithms.

f. Deployment and Integration:

Install the system at selected traffic intersections in a real-world urban environment.
Ensure seamless integration with existing traffic management infrastructure.

6|Page
Facilities required for proposed work.
To develop the proposed Dynamic Traffic Light Controller System, you'll need a range of software and
hardware resources. Here's a breakdown of the required facilities, software, and hardware.

Facilities:

1. Workspace: A dedicated workspace with desks, chairs, and sufficient space for project planning,
development, and testing.

2. Meeting and Collaboration Space: Facilities for team meetings, discussions, and collaboration
with project stakeholders.

3. Data Collection Locations: Access to real-world traffic intersections for data collection and system
testing (with permission from relevant authorities).

Software:
1. Development Tools:

- Integrated Development Environments (IDEs) like Visual Studio Code, PyCharm, or Eclipse for software
development.

- Version control systems (e.g., Git) for code management and collaboration.

2. Data Processing and Analysis:

- Python or R for data processing and analysis.

- Libraries like NumPy, pandas, Matplotlib, and seaborn for data manipulation and visualization.

3. Machine Learning and Deep Learning:

- TensorFlow, PyTorch, or Keras for developing and training machine learning and deep learning models.

4. Traffic Signal Control Algorithms:

- Software development tools and languages for implementing traffic signal control algorithms, which may
include Python, Java, or C++.

5. Simulation and Testing Tools:

- Tools for simulating traffic scenarios and evaluating the performance of the traffic signal optimization
system.

7|Page
Hardware:

1. Computers and Servers:

- High-performance computers or servers for data processing, machine learning model training, and hosting
the system.

2. Networking Infrastructure:
- Reliable network connections for data transmission and communication between system components.

3. Storage Solutions:
- Sufficient storage capacity for storing historical traffic data and system logs.

4. Testing and Validation Equipment:

- Equipment for field testing and validation of the system's performance at real traffic intersections.

8|Page
References for this Project:
Certainly, here are some references for our project on dynamic traffic light controller
systems and related topics:

Traffic Signal Control and Optimization:

1. Ritchie, S. G., & Tomlin, C. J. (2001). The Use of Traffic Signal Control for
Combating Congestion. Transportation Research Part B: Methodological, 35(5), 383-398.

2. Papageorgiou, M., & Kotsialos, A. (2002). Review of Road Traffic Control Strategies. IEE
Proceedings - Control Theory and Applications, 149(5), 352-366.

SCAT (Sydney Coordinated Adaptive Traffic System):

3. McCloskey, M. V., & Tarko, A. P. (2005). Real-Time Traffic Signal Control with
Sydney Coordinated Adaptive Traffic System. Journal of Transportation Engineering, 131(6), 466-474.

4. Lustig, I. J., & Luh, P. B. (1995). An Adaptive Traffic Signal Control System:
Implementation Issues and Impacts. Transportation Research Part C: Emerging Technologies, 3(4), 229-249.

ATCS (Adaptive Traffic Control Systems):

5. Apon, A. W., & Sun, C. (2015). Adaptive Traffic Signal Control Systems: A
Literature Review. Journal of Transportation Engineering, Part A: Systems, 141(5), 04015001.

6. Gartner, N. H., & Messer, C. J. (2019). A Review of Adaptive Traffic Signal Control Methods.
Transportation Research Part C: Emerging Technologies, 107, 111-128.

SCOOT (Split Cycle Offset Optimization Technique):

7. Robertson, D. I., & Bretherton, R. D. (1994). SCOOT: A Traffic Responsive Method of

Coordinating Signals. Traffic Engineering & Control, 35(3), 158-161.

Genetic Algorithms in Traffic Signal Optimization:

8. Gao, Y., & Zheng, D. (2019). Genetic Algorithms for Traffic Signal Timing Optimization: A
Review. Journal of Advanced Transportation, 2019, 9610719.

9|Page
Deep Neural Networks in Traffic Signal Optimization:

9. Li, S., & Sun, Y. (2021). Deep Reinforcement Learning for Traffic Signal Control: A Review.
Transportation Research Part C: Emerging Technologies, 132, 385-416.

10. Zhang, Z., & Wu, J. (2019). Deep Reinforcement Learning for Traffic Signal
Control in Smart Cities: Recent Advances and Future Challenges. IEEE Transactions on Intelligent
Transportation Systems, 21(9), 3690-3700.

10 | P a g e