Real-Time Estimation of Queue at Signalized Intersection Using RFID Sensors Under Mixed Traffic Conditions

This chapter delves into the innovative use of RFID sensors for real-time queue estimation at signalized intersections, addressing the challenges posed by mixed traffic conditions. Traditional loop detectors, while effective in homogeneous traffic, fall short in environments like India, where lane discipline is often lacking, and vehicle heterogeneity is high. The study introduces RFID sensors as a non-intrusive and efficient alternative, leveraging the RFID tags mandated on four-wheelers sold in India post-2014. The research focuses on two primary scenarios: Queue Within Detector (QWD) and Queue Beyond Detector (QBD), each presenting unique challenges and requiring distinct estimation models. The chapter meticulously outlines the data collection process, including the strategic placement of RFID sensors and the synchronization of data from multiple sources. It then presents the development and validation of queue estimation models using Kalman filter and extended Kalman filter techniques, demonstrating their superior accuracy compared to traditional methods. The findings underscore the potential of RFID sensors and advanced filtering techniques in enhancing traffic management strategies, particularly in complex, mixed traffic environments. The chapter concludes with a discussion on the implications for intelligent transportation systems and suggests avenues for future research, making it a compelling read for those interested in cutting-edge traffic management solutions.