Researchers from the Department of Electronics and Communication Engineering at the National Institute of Technology (NIT) Rourkela have developed a smart roadside system that can detect vehicles approaching blind corners and alert drivers in real time.

The system uses surveillance cameras integrated with Internet of Things (IoT) and edge computing technology to monitor sharp turns. Instead of sending video data to distant cloud servers, the framework processes information locally through on-site edge devices. This significantly reduces communication delays and enables instant audio-visual warnings for approaching drivers.

The research, published in the International Journal of Computational Vision and Robotics, was co-authored by BTech graduate K.L. Sanjeev Tudu along with Prof. Santos Kumar Das, Prof. Umesh Chandra Pati, Prof. Poonam Singh, Dr. Goutam Kumar Sahoo, and Dr. Rashmiranjan Nayak.

In real-world tests, the system successfully detected vehicles in blind zones and accurately estimated their speed and distance, even on low-power devices. The framework also includes a graphical interface for traffic management centres to monitor high-risk intersections.

In a related development, the team has secured a patent for a low-cost IMU/GPS-based real-time accident detection and alert system built using ESP32 technology. Designed at an estimated one-time cost of ₹6,000, the system uses embedded sensors and GPS to detect accidents and automatically alert emergency responders. A companion smartphone application, “Track,” enables crash victims to quickly connect with emergency contacts and authorities.

The researchers are now working on integrating roadside vision systems with embedded accident detection units to create a hybrid AI-based safety network aimed at enhancing smart city and connected mobility infrastructure, particularly in rural and hilly regions.