Problem Statement
One problem these traffic lights systems fail to accommodate is to adapt to emergency situations like emergency vehicles passing through. They are not intended to adapt when emergency vehicles, such as ambulances, fire trucks, or police cars, need to pass through intersections where their lane is halted. These systems typically operate on fixed, pre-programmed schedules and are not equipped to detect or respond to the presence of emergency vehicles. As a result, emergency vehicles may be forced to wait at red lights, wasting crucial time that could otherwise be spent reaching their destinations. In situations where seconds can mean the difference between life and death, these delays can have serious consequences that could impact the ability of first responders to save lives or deal with dangerous situations.
Another problem we identified is traffic light systems, even when enhanced with machine learning, often fail to account for the impact of varying weather conditions on traffic flow. Weather events such as heavy rain, snow, fog, and extreme temperatures can significantly alter driving behavior, reduce road visibility, and slow down vehicle speeds, leading to congestion and increased accident risk. Despite these factors, many existing traffic management systems rely on fixed timing schedules or traffic data that do not incorporate real-time weather information. As a result, traffic lights may remain inefficient during adverse weather conditions, exacerbating delays and road safety concerns. This research aims to explore how machine learning models can be designed to dynamically adapt traffic signal control based on real-time weather data to enhance traffic flow and safety under such conditions.