International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 08 | Aug 2023
p-ISSN: 2395-0072
www.irjet.net
Smart Traffic Congestion Control System: Leveraging Machine Learning for Urban Traffic Optimization P. Venkata Srinivasa Reddy1, A. Bhavani2, Ch. Saranya3 1 Student, Dept of AI & DS, VVIT, Andhra Pradesh, India 2Student, Dept of IT, VVIT, Andhra Pradesh, India
3Student, Dept of IT, VVIT, Andhra Pradesh, India ---------------------------------------------------------------------***--------------------------------------------------------------------inputs and manual calculations. The heart of the congestion Abstract - Urban traffic congestion poses a significant
detection process is a Convolutional Neural Network (CNN) model, trained on a diverse dataset comprising over 1000 CCTV monitoring images. In a time of urban expansion and growing traffic demands, the integration of these innovative systems is pivotal for developing an efficient, eco-friendly transportation network. Such a network not only benefits commuters by reducing travel times and enhancing safety but also contributes to a cleaner environment by minimizing unnecessary fuel consumption and emissions.
challenge, leading to extended travel times, heightened pollution, and mounting frustration. To combat this issue, we propose the introduction of a Smart Traffic Congestion Control system, which leverages technology to optimize traffic flow. Our objective is to design an intelligent traffic system that dynamically adjusts signal timings using real-time data analysis and predictive modelling. To achieve this, we are integrating advanced machine learning technologies such as Proximal Policy Optimization (PPO), Long Short-Term Memory (LSTM), and YOLOv4, for facilitating timely decisionmaking for improved traffic patterns and capturing intricate traffic behaviour. By harnessing data-driven decision-making and intelligent algorithms, the smart congestion control system has the potential to revolutionize traffic control strategies, offering a sustainable and efficient approach to urban mobility. In the context of rapidly growing cities and escalating traffic demands, the implementation of such advanced systems becomes imperative for establishing a seamless and eco-friendly transportation network that benefits both commuters and the environment.
This System explores the architecture, design, and performance of the Traffic Congestion Control System, shedding light on its potential to revolutionize traffic management in urban environments. We demonstrate how the fusion of machine learning, real-time data analysis, and predictive modelling can pave the way for a smarter, more sustainable future in urban transportation.
2. LITERATURE REVIEW Traffic congestion in urban areas is a multifaceted problem that has persisted for decades, leading researchers and engineers to explore innovative solutions to alleviate its adverse effects on society, the environment, and the economy. The integration of Machine Learning (ML) and artificial intelligence (AI) techniques into traffic management systems has emerged as a promising avenue to tackle this challenge effectively. This literature review provides an overview of relevant studies and developments in the field of ML-based urban traffic management.
Key Words: Machine Learning, YOLOv4, LSTM, PPO, Traffic Congestion
1. INTRODUCTION Urban traffic congestion poses a significant challenge to transportation systems worldwide, leading to increased commute times, environmental pollution, and economic losses. In response to this pressing issue, we introduce a cutting-edge Traffic Congestion Control System that harnesses the power of Machine Learning to transform urban traffic management.
2.1 Traffic Congestion Challenges: Traffic congestion is a complex issue influenced by factors such as population growth, urbanization, and the increasing number of vehicles on the road. It leads to significant economic losses, increased fuel consumption, and heightened levels of air pollution. Traditional traffic management systems have often fallen short in addressing these challenges.
This system combines a range of advanced technologies to achieve its objectives, with a primary focus on optimizing signal timings at intersections and interconnected routes. By utilizing real-time data from live cameras installed at traffic points, it dynamically allocates signal durations to mitigate congestion effectively.
2.2 Machine Learning for Congestion Detection:
A key innovation lies in the application of deep learning techniques for congestion detection. To improve efficiency, the system employs preprocessing methods for smaller camera images, reducing the dependency on high-quality
© 2023, IRJET
|
Impact Factor value: 8.226
Researchers have increasingly turned to ML algorithms for traffic congestion detection. One of the key contributions of
|
ISO 9001:2008 Certified Journal
|
Page 874