Source-linked AI summary
Optimal Control of Connected and Automated Vehicles at Roundabouts: An Investigation in a Mixed-Traffic Environment
Liuhui Zhao, Andreas Malikopoulos, Jackeline Rios-Torres
TL;DR
Roundabouts can become congestion bottlenecks, motivating research into coordinating connected and automated vehicles to improve traffic flow. The paper analytically optimizes vehicle acceleration and deceleration and evaluates coordination through simulation. Fully coordinated vehicles yielded 51% travel time savings, while improvements under near-capacity mixed traffic were less substantial.
Problem
The paper examines how to optimally coordinate CAVs entering roundabouts to improve traffic flow and how coordination affects fuel consumption and travel time across market penetration levels.
Source: §1 · p. 2Method
Results
Takeaways & Limitations
Under the studied near-capacity demand, mixed traffic with high CAV penetration still performed less substantially than a fully coordinated CAV network.
Source: §5 · p. 6Under high demand, a stopped non-CAV can readily form a queue that is difficult to dissolve, limiting mixed-traffic benefits even at high CAV penetration.
Source: §4 · p. 5
Abstract
from arXiv · showhide
Connectivity and automation in vehicles provide the most intriguing opportunity for enabling users to better monitor transportation network conditions and make better operating decisions to improve safety and reduce pollution, energy consumption, and travel delays. This paper investigates the implications of optimally coordinating vehicles that are wirelessly connected to each other in roundabouts to achieve a smooth traffic flow. We apply an optimization framework and an analytical solution that allows optimal coordination of vehicles for merging in such traffic scenario. The effectiveness of the proposed approach is validated through simulation and it is shown that fully coordinated vehicles reduce total travel time by 51% and fuel consumption by 35%. Furthermore, we study the influence of vehicle coordination in a mixed-traffic environment and compare the network performance under different market penetration rates of connected and automated vehicles (CAVs). For this particular study with near-capacity demand, due to extremely unstable traffic, the results show that even with high penetration of CAVs (e.g., 80%), travel time and fuel savings are much less than a network of CAVs.
1. INTRODUCTION
The introduction frames roundabouts as potential congestion bottlenecks and motivates vehicle coordination as a way to improve traffic flow. The paper formulates optimal CAV coordination at roundabouts and investigates its effects on fuel consumption and travel time across market penetration levels.
1. INTRODUCTION
- Intersections, roundabouts, merging roadways, and speed-reduction zones are identified as bottleneck sources contributing to traffic congestion.The cited estimate for 2015 reports 6.9 billion extra hours on the road and 3.1 billion extra gallons of fuel purchased in US urban areas.Source: §1 · p. 1
- Roundabouts can reach capacity even under moderate demand, and heavy circulating flow can significantly delay minor-road approaches.Prior work discussed in the introduction focuses mainly on metering and traffic-signal controls to improve roundabout operations.Source: §1 · pp. 1–2
- Prior CAV coordination research addressed merging and intersections, including vehicle control, safe headways, platooning, and reservation schemes.The introduction positions the present work among efforts to improve traffic flow through coordination at different transportation segments.Source: §1 · p. 2
- The study formulates optimal coordination of CAVs entering roundabouts and investigates fuel consumption and travel time at different market penetration levels.Its contributions include an analytical solution for each vehicle’s optimal acceleration and deceleration, evaluated through simulation under different traffic conditions.Source: §1 · p. 2
2. PROBLEM FORMULATION
The formulation models CAVs entering a single-lane roundabout from two freeways, where a control zone precedes the merging zone. It coordinates their entry times by minimizing vehicle control effort subject to vehicle, speed, and collision-safety constraints.
2. PROBLEM FORMULATION
- Vehicles from two freeways approach a single-lane roundabout through a control zone; the merging zone is the arc where lateral collisions may occur.The coordinator communicates with vehicles in the control zone but does not decide vehicle actions.Source: §2 · p. 2
- Each vehicle is modeled with position and speed states and acceleration or deceleration as its control input, using double-integrator dynamics.Vehicles are ordered in a first-in-first-out queue, with control and speed constrained to admissible ranges.Source: §2 · pp. 2–3
- The problem minimizes each vehicle’s acceleration/deceleration input from control-zone entry to exit, subject to a hard rear-end safety constraint.The model also imposes admissible acceleration and speed ranges and assumes vehicles cruise at the roundabout speed limit.Source: §2 · p. 3
- A coordinator assigns vehicles FIFO identities and relays exit-time information so each following vehicle can compute its own control-zone exit time.The vehicle’s merging-zone entry time can be computed from its control-zone exit time.Source: §2 · p. 3
- Same-road followers must preserve a predefined safe distance, while vehicles from different roads may enter the merging zone only after the preceding vehicle exits.The constraints prevent rear-end and lateral collisions, respectively.Source: §2 · p. 3
3. SIMULATION FRAMEWORK
The study implements a closed-form vehicle-control solution in VISSIM and evaluates a roundabout simulation under varied CAV penetration rates. The setup specifies the network, vehicle behavior, demand, and performance measurements.
3. SIMULATION FRAMEWORK
- The closed-form solution is implemented through the VISSIM API, which updates vehicle acceleration or deceleration every 0.05 seconds using collected vehicle information.Trajectories are archived every second, while aggregated travel-time and delay data are recorded every 60 seconds.Source: §3 · pp. 3–4, Figure 2 · p. 4
- The VISSIM network has one merging point: westbound vehicles circulate while eastbound vehicles merge, with 12 m merging zones and 300 m control zones per approach.The roundabout perimeter is approximately 200 m, and each input point is 320 m from the roundabout entry.Source: §3 · p. 4, Figure 3 · p. 4
- The single-lane roundabout uses 15.6 m/s entry speeds, an 8.9 m/s internal speed limit, and automated-vehicle acceleration up to 4.5 m/s².For general-purpose vehicles, the stated maximum acceleration and minimum deceleration settings are 3.0 m/s² and -4.5 m/s².Source: §3 · p. 4
- CAV market penetration rates of 0%, 20%, 50%, 80%, and 100% are tested; CAV control switches off when the leading non-CAV activates the safety constraint.At 0% CAV penetration, the study applies the Wiedemann car-following model to non-CAVs.Source: §3 · p. 4
- Five simulation runs evaluate a near-capacity demand of 800 vehicles per hour per lane, with 400 vehicles dispatched over 900 seconds and 1,200 seconds of simulation.Total travel time and fuel consumption are summed over dispatched vehicles; travel time and delay are collected as performance measures.Source: §3 · p. 4
4. RESULTS
At 100% CAV market penetration, coordination smooths traffic and improves network travel time and fuel consumption; under mixed traffic near capacity, benefits are less substantial.
4. RESULTS
- At 0% MPR, near-capacity demand makes eastbound merging difficult, building a queue until vehicle input stops at 900 seconds.Westbound density remains stable; congestion gradually releases after arrivals stop.Source: §4 · p. 4, Figure 4 · p. 5
- At 100% MPR, eastbound density becomes stable and the roundabout serves more cumulative vehicles in the same period than at 0% MPR.Coordination lets eastbound vehicles merge without stops despite high circulating flow.Source: §4 · p. 5, Figure 4 · p. 5, Figure 5 · p. 5
- 51% travel time savings is observed for the entire network under 100% MPR, alongside fuel-consumption savings.Coordination minimizes variation between the two approaches and eliminates eastbound stop-and-go driving.Source: §4 · p. 5, Figure 7 · p. 5, Figure 8 · p. 5
- In mixed traffic, even high CAV penetration such as 80% yields only modest travel-time and fuel-consumption benefits under near-capacity demand.A stopped leading non-CAV can trigger a queue that is difficult to dissolve under high circulating flow, delaying following CAVs.Source: §4 · p. 5, Table 1 · p. 6
5. CONCLUSIONS
Simulation results show that vehicle coordination improves travel time and fuel consumption under 100% CAV market penetration. For near-capacity demand, the improvement is less substantial in mixed traffic, and future work will examine other traffic settings and roundabout designs.
5. CONCLUSIONS
- Simulation results show significantly improved travel time and fuel consumption under 100% CAV market penetration.For near-capacity demand, improvement under mixed traffic is less substantial than under 100% CAV penetration.Source: §5 · p. 6
- The reported efficiency evaluation uses simulation of CAVs controlled before entering a roundabout to form smooth traffic flow.Source: §5 · p. 6
- Future work will investigate CAV and non-CAV interactions and improve control at low CAV market penetration rates.The paper also identifies multi-lane roundabouts with more conflict points and weaving sections as a future research setting.Source: §5 · p. 6