In recent years, large-scale testing has begun for connected and autonomous driving, making it possible to implement the concept of Dynamic Lane Reversal (DLR), which can quickly shift lane directions to reflect instantaneous flow dynamics. DLR is to make full use of road space and avoid waste of road capacity, and potentially alleviate congestion. However, the impact and feasibility of DLR remain unclear. In order to investigate the effectiveness, the feasibility, and the applicability of DLR, we utilize a direction-changeable, lane-based cell transmission model to find an optimal DLR scheme for a roadway segment with stochastic traffic flow in both directions. A proxy model is also designed to realize DLR in VISSIM. A regression analysis was carried out to find the impacts of directional flow rate and the number of lanes on delay reductions due to DLR. Results indicate that implementing DLR can reduce the total queuing delay considerably compared to traditional reversible lane strategies. We found that DLR achieved superior performance on segments with more lanes and when the flows from each direction were close to one another. A zigzag frontier of the delay reduction was discovered. Findings from this research shed light on the feasibility and effectiveness of DLR on various types of road segments.
Modeling and simulation of dynamic lane reversal using a cell transmission model
Journal of Intelligent Transportation Systems ; 26 , 6 ; 717-729
2022-10-21
13 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Unbekannt
A cell transmission model for dynamic lane reversal with autonomous vehicles
Online Contents | 2016
|Dynamic Lane Reversal: A reinforcement learning approach
IEEE | 2023
|Max-pressure control of dynamic lane reversal and autonomous intersection management
Taylor & Francis Verlag | 2019
|