In a recent study, the International Journal of Information and Communication Technology unveiled a new approach for managing urban traffic across multi-intersection networks that could significantly reduce traffic congestion and enhance road efficiency.
Researchers from the Chongqing Institute of Engineering in China suggested the use of vehicle-to-everything (V2X) technology to enable real-time data exchange between vehicles and infrastructure, providing critical insights on road conditions and traffic.
Leveraging Modern Technologies
The proposed initiative aims to improve traffic management systems, enhancing the control of traffic lights, speed, and lane restrictions to ensure a smoother and safer flow of vehicles.
The system leverages an optimized long short-term memory (LSTM) model, a sophisticated artificial intelligence (AI) designed to recognize patterns and generate predictions.
Critical to the system’s effectiveness is the “sliding time window” update mechanism, which allows the model to learn from real-time data while maintaining historical context. This innovative approach enables rapid adjustments to traffic flow while reducing the computational load, cutting prediction times in half.
Simulations of this strategic approach demonstrated a reduction in the average vehicle delays by one-third and road throughput increased by almost 15%. These advancements permit shorter travel times, smoother traffic flow, improved fuel consumption, and reduced vehicle emissions.
Shifting from Conventional Traffic Methods
The study emphasized that the system delivers a substantial advantage over conventional traffic management methods, which rely on historical data or limited real-time inputs, resulting in untimely responses to actual road conditions.
Moreover, while conventional traffic systems are valuable in less complex traffic situations, they struggle to handle rapid and unpredictable changes in traffic, particularly in larger and interconnected networks.
As cities worldwide grapple with the growing traffic congestion, the newly proposed system aims to address the limitations of conventional systems to deliver more innovative, responsive, and precise solutions for traffic management.