Scholarlink

Traffic congestion is a pressing issue in urban areas, especially in India, where road intersection topologies, road structures, and traffic coordination among intersections impose challenges in traffic routing and ensuring road safety. It results in significant fuel wastage, elevated CO2 emissions, stress among the road users, and diminished individual productivity, thereby impacting the overall GDP. Traditional traffic management systems struggle to adapt to the highly dynamic and unstructured nature of Indian road conditions. There is a critical need for innovative solutions that can effectively address these challenges. Dynamic traffic management using the reinforcement learning approach is widely adopted and greatly supports resolving issues in managing real-time traffic effectively. Strategies to minimize traffic congestion at multi-type intersections have been insufficiently explored in recent research. To overcome this scenario, we present microscopic traffic flow analysis for multi-type intersection roads like four-leg intersections, ramp merges, Y-type, T-type, and roundabouts using the Simulation of Urban Mobility (SUMO) traffic simulator. Also, reinforcement learning (RL) based strategies are formulated to manage the traffic congestion occurring in identified metropolitan cities, including Chennai, Madurai, Tiruvallur, Coimbatore, and Trichy. This study will help researchers gain insights into the characteristics of real-world road environments, and its findings will serve as valuable support for the development of advanced dynamic traffic management systems.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Streamlining Traffic Flow with Smart Mobility for Vehicles Through Reinforcement Learning Based Framework

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Santhiya Ravindran,
  • Varun Prakash Rajamohan,
  • Senthil Kumar Jagatheesaperumal

摘要

Traffic congestion is a pressing issue in urban areas, especially in India, where road intersection topologies, road structures, and traffic coordination among intersections impose challenges in traffic routing and ensuring road safety. It results in significant fuel wastage, elevated CO2 emissions, stress among the road users, and diminished individual productivity, thereby impacting the overall GDP. Traditional traffic management systems struggle to adapt to the highly dynamic and unstructured nature of Indian road conditions. There is a critical need for innovative solutions that can effectively address these challenges. Dynamic traffic management using the reinforcement learning approach is widely adopted and greatly supports resolving issues in managing real-time traffic effectively. Strategies to minimize traffic congestion at multi-type intersections have been insufficiently explored in recent research. To overcome this scenario, we present microscopic traffic flow analysis for multi-type intersection roads like four-leg intersections, ramp merges, Y-type, T-type, and roundabouts using the Simulation of Urban Mobility (SUMO) traffic simulator. Also, reinforcement learning (RL) based strategies are formulated to manage the traffic congestion occurring in identified metropolitan cities, including Chennai, Madurai, Tiruvallur, Coimbatore, and Trichy. This study will help researchers gain insights into the characteristics of real-world road environments, and its findings will serve as valuable support for the development of advanced dynamic traffic management systems.