Traffic management with autonomous and connected vehicles at single-lane roundabouts. (April 2021)

• Record Type: Journal Article
• Authors: Martin-Gasulla, Marilo; Elefteriadou, Lily
• Is Part Of: Transportation research. Volume 125(2021)
• Publication Date: 2021-04
• Subjects: Roundabout -- Roundabout manager -- Autonomous and connected vehicles -- Control delay -- Roundabout's capacity -- Expert systems; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2021.102964
• Languages: English
• ISSNs: 0968-090X

Abstract

Abstract: The continuous evolution of automotive and communication technologies can result in more efficient traffic management. At roundabouts, Autonomous and Connected Vehicles (CAVs) have the potential to enhance traffic operations, minimizing lost time and providing higher capacities by improving the gap acceptance process. This research aims to take advantage of those evolving technologies in order to design a more efficient traffic management system for single-lane roundabouts. Such a system would provide optimal coordination of CAVs to negotiate the roundabout, maximizing throughput and minimizing average control delay. The expert system developed and tested (Roundabout Manager) acts as a central controller and manages conflicts within the roundabout. It prioritizes incoming vehicles based on demand-dependent strategies and adjusts their trajectories accounting for the geometry of the roundabout. Seven strategies are proposed to coordinate CAV trajectories. The first strategy prioritizes incoming vehicles based on their Shortest-Remaining-Time-First (SRTF, i.e., the shortest time till the vehicle reaches the first conflict section it will encounter) to the conflict sections within the circular roadway, while the other strategies add to the SRTF rules to accommodate highly unbalanced traffic flows, with heavy right- or left-turning maneuvers. To illustrate the operation of the algorithm and assess its effectiveness as a function of the demand distribution, a total of 15 oversaturated demand scenarios with 10 replications per scenario were simulated for each of the seven strategies examined. It was concluded that, under any of the strategies proposed, the system guarantees higher throughput with lower average control delay compared to conventional vehicles' operation. The capacity of the roundabout under fully CAVs traffic flow is increased by 58 to 73% (from low to high conflicting flow, respectively). The strategy that prioritizes vehicles that have to travel through more conflict points and those that leave the system first outperforms the other strategies at all demand scenarios, becoming the optimal coordination of CAVs at roundabouts. This strategy reduces the average control delay between 80 and 97% compared to traffic with conventional vehicles, as assessed using the Highway Capacity Manual procedures.