A ROLLBACK 1 APPROACH FOR DEMAND CONSISTENCY CHECKING OF REAL-TIME TRAFFIC NETWORK STATE ESTIMATION MODELS

摘要

The paper presents a real-time traffic network state estimation model with on-line demand consistency checking and updating capabilities. Different from reactive-based methodologies proposed in the literature, the model adopts a time rollback with corrective actions approach. If an instance of inconsistency between the measured and estimated network state is observed, the model is allowed to rollback in time and promptly re-simulate a pre-defined past period after adjusting the appropriate model's parameters to minimize the observed inconsistency. A demand correction algorithm is developed and used for demand adjustment for each rollback period. The results of applying the developed model for a test-bed network are presented. The results show that the approach improves the model's consistency with real-world observations.