Scholarlink

This research presents a Dynamic Traffic Control System (DTCS) that leverages IoT and Machine Learning to optimize urban traffic signal management. The proposed system operates in two phases. In the first phase, it prioritizes emergency vehicles by equipping them with ESP8266 modules that communicate with ESP32 receivers at intersections. A Raspberry Pi controller processes this data and adjusts signals in real time to ensure smooth emergency transit. In the second phase, green light durations are dynamically adjusted based on real-time traffic volume captured via cameras, with processing handled by a lightweight machine learning model, VDDNet (Vehicle Density Detection Network). Experimental evaluations and SUMO simulations with 300 vehicles showed a 5.87% reduction in total traffic time, 5.88% in average travel time, and 4.95% in average vehicle wait time. The system is cost-effective, scalable, and built using commercially available IoT hardware, making it a practical alternative to conventional traffic management solutions in smart cities.

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

IoT-Enabled Emergency Vehicle Detection and Adaptive Traffic Signal Control Using Custom Object Detection and Dynamic Time Allocation

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Shubham Pawar,
  • Avi Poptani,
  • Manan Shah,
  • Vishal Barot

摘要

This research presents a Dynamic Traffic Control System (DTCS) that leverages IoT and Machine Learning to optimize urban traffic signal management. The proposed system operates in two phases. In the first phase, it prioritizes emergency vehicles by equipping them with ESP8266 modules that communicate with ESP32 receivers at intersections. A Raspberry Pi controller processes this data and adjusts signals in real time to ensure smooth emergency transit. In the second phase, green light durations are dynamically adjusted based on real-time traffic volume captured via cameras, with processing handled by a lightweight machine learning model, VDDNet (Vehicle Density Detection Network). Experimental evaluations and SUMO simulations with 300 vehicles showed a 5.87% reduction in total traffic time, 5.88% in average travel time, and 4.95% in average vehicle wait time. The system is cost-effective, scalable, and built using commercially available IoT hardware, making it a practical alternative to conventional traffic management solutions in smart cities.