Experimental Validation of a Real-Time Optimal Controller for Coordination of CAVs in a Multi-Lane Roundabout (2001.11176v3)

Abstract: Roundabouts in conjunction with other traffic scenarios, e.g., intersections, merging roadways, speed reduction zones, can induce congestion in a transportation network due to driver responses to various disturbances. Research efforts have shown that smoothing traffic flow and eliminating stop-and-go driving can both improve fuel efficiency of the vehicles and the throughput of a roundabout. In this paper, we validate an optimal control framework developed earlier in a multi-lane roundabout scenario using the University of Delaware's scaled smart city (UDSSC). We first provide conditions where the solution is optimal. Then, we demonstrate the feasibility of the solution using experiments at UDSSC, and show that the optimal solution completely eliminates stop-and-go driving while preserving safety.

• The paper experimentally validates a real-time optimal control framework for coordinating connected and automated vehicles (CAVs) in multi-lane roundabouts using a scaled smart city testbed.
• The framework models CAVs using a double integrator, formulating the problem to minimize energy and speed variations while ensuring safety constraints and collision avoidance.
• Experimental results demonstrate the elimination of stop-and-go driving and highlight the potential for integrating optimized control strategies into future urban CAV systems.

Analyzing Optimal Control for CAV Coordination in Multi-Lane Roundabouts

This paper presents a comprehensive paper on the implementation of an optimal control framework for managing Connected and Automated Vehicles (CAVs) in multi-lane roundabouts. The research leverages the capabilities of the University of Delaware's scaled smart city (UDSSC) testbed to experimentally validate a previously developed decentralized optimal control framework aimed at real-time coordination of CAVs. The paper specifically addresses the eliminations of stop-and-go driving patterns, which are critical for enhancing vehicle efficiency and network throughput in roundabout scenarios.

Framework and Problem Formulation

The optimal control framework is modeled around a multi-lane roundabout with distinct inflow and collision areas, offering a structured environment wherein CAVs dynamically interact. The crux of their approach is in minimizing the spatial and temporal speed variations across the vehicles, thereby smoothing traffic flow and preventing stop-and-go waves.

The authors formulated their problem by using a double integrator model for individual CAVs under a set of constraints that include speed limits and safety distances. The problem emphasizes energy minimization during navigation through the roundabout's control zone, balancing this objective against conditions that ensure lateral and rear-end collision avoidance.

Analytical and Experimental Insights

In the analytical phase of the research, the authors derive an unconstrained optimal trajectory leveraging Hamiltonian analysis, allowing each CAV to determine its path through the roundabout by adjusting its timeline based on initial speed and control zone parameters. Despite the traditional challenges posed by state and control constraints, the chosen formulation ensures feasibility by focusing on time optimization within calculated safety constraints.

The experimental setup at the UDSSC testbed involved implementing this trajectory optimization for nine CAVs, verifying the system’s efficacy across several runs. One important conclusion from these trials is the elimination of stop-and-go driving, with results indicating that the minimum speed of any CAV in the scaled setup well exceeds baseline expectations for smooth automotive flow. Such outcomes underscore the potential of optimized control strategies in future vehicular systems wherein automation and connectivity are standard.

Implications and Future Work

The implications of this paper are significant at both practical and theoretical levels. Practically, the ability to handle dynamic multi-lane roundabout scenarios can inform better design and integration of CAVs into urban environments, particularly in areas where traditional traffic signals and signs are less effective. Theoretically, the demonstration of the system's feasibility under the constraints outlined contributes to the broader field of optimal control in automated transport networks.

Future research should aim to incorporate and mitigate potential variability factors including communication delays and tracking errors. The practical extension of this framework should consider large-scale deployment challenges, assessing trade-offs in control flexibility and safety constraints to optimize both for efficiency and reliability in real-world applications.

While the framework presented offers a robust solution to roundabout coordination for CAVs, its integration with existing infrastructure and adaptation to diverse vehicular dynamics across different urban layouts remains an open field for exploration. Such investigations are essential for advancing the state-of-the-art in autonomous vehicle control systems, propelling us toward smarter and more sustainable urban mobility solutions.