Real-time Vehicle Reidentification System for Freeway Performance Measurements

Computational resources in the traffic operation field as well as the bandwidth of field communication links, are often quite limited. Accordingly, for real-time implementation of Advanced Transportation Management and Information Systems (ATMIS) strategies, such as vehicle reidentification, there is strong interest in development of field-based techniques and models that can perform satisfactorily while minimizing computational and communication requirements in the field. The ILD (Inductive Loop Detector)-based Vehicle ReIDentification system (ILD-VReID) is an example of a currently applied approach. Although ILDs are not without limitations as a traffic sensor, they are widely used for historical reasons and the sunken investment in the large installed base makes their use in this research highly cost-effective. Therefore, this dissertation develops a new vehicle reidentification algorithm, RTREID-2, for real-time implementation by adopting a PSR (Piecewise Slope Rate) approach that extracts features from raw vehicle signature data. The results of cases studies indicate that RTREID-2 is capable of accurately providing individual vehicle tracking information and performance measurements such as travel time and speed. The potential contributions of RTREID-2 are: application to square and round single loop configurations, and reduced computational requirements because re-estimation or transferability of the speed models used in the previously developed approach is not necessary. As a consequence, RTREID-2 is free of site-specific calibration and transferability issues. A freeway corridor study also demonstrates that RTREID-2 has the potential to be implemented successfully in a congested freeway corridor, utilizing data obtained from both homogenous and heterogeneous loop detection systems. A real-time vehicle classification model, which is based on the PSR approach, was also developed on the part of RTREID-2. The classification model can successfully classify vehicles into 15 classes using single loop detector data without any explicit axle information. The initial results also suggest the potential for transferability of the vehicle classification approach and are very encouraging. To investigate real-time freeway performance measurement in a realworld setting, the design of a RTPMS (Real-time Traffic Performance Measurement System) that is based on RTREID-2 is also presented in this dissertation. A simulation of RTPMS is conducted to evaluate its feasibility. The simulation results demonstrate the potential of implementing RTPMS in real world applications.