Distributed Architecture and Algorithm for Robust Real-Time Progression Evaluation and Improvement (0-4729-2)

Closed-Loop systems are widely implemented in Texas arterials to provide efficient operation of arterial intersections while still providing signal progression. Nevertheless, poor progression can be observed along most arterials due to outdated offsets, short-term variations in traffic patterns (early-return-to-green), or changes in arterial's speed and changes in traffic volumes. The limited abilities of closed-loop systems to adapt to traffic variations have stimulated interest into incorporating the technologies of adaptive control software (ACS) into closed-loop systems in order to address such issues. These integrated-type systems require lower cost and minimal staff training, in comparison to fully adaptive systems, since traffic engineers and technicians managing the traffic signal operating systems are already familiar with the closed-loop logic. The objective of this research is to develop, implement, and test an algorithm that will address the limitations of previous efforts in the area of real-time offset-tuning. This final report describes a flexible experimental framework for theoretical development and experimentation of adaptive control algorithms, in the context of closed-loop systems operation. The developed system includes: 1) robust classification algorithm of progression quality and remedies, and 2) vendor-independent distributed implementation architecture for the proposed algorithm.