Machine Learning-Based Dynamic Signal Switching: A Promising Solution to Traffic Congestion

Today's growing urban population means increased revenue for individuals, translating into growing vehicles on the road. Nowadays, people use their private vehicles instead of public transportation, which results in a considerable number of personal vehicles that lead totraffic congestion everywhere. The ever-increasing number of vehicles and complex congestion leads to frequent traffic jams in major cities. In today's scenario, traditional traffic management methods have become redundant as the traffic volume has grown exponentially. The primary goal of this proposed work is to develop an algorithm that uses machine learning to control traffic effectively and ensure efficient traffic flow in the event of unforeseen events like accidents. The existing algorithm focuses on predicting shortterm traffic flow but ignores that traffic congestion may also occur due to accidents and emergencies. Our proposed model will consider all the above factors and use dynamic signal switching among lanes to predict the best and most efficient solution to reducing waiting time for vehicles and congestion on the road. To create an effective traffic flow management model with the least amount of road mayhem, our suggested model needs traffic data as input, which was then trained using various Machine Learning (ML) models.