TLDR: A new framework called HCOMC has been developed to improve highway on-ramp merging in mixed traffic (human-driven and automated vehicles). It uses hierarchical planning, game theory for lane changes, and multi-objective optimization to enhance safety, stability, traffic efficiency, and fuel economy, outperforming existing control models in simulations.
Highway on-ramp merging areas are notorious for causing traffic jams and accidents. The challenge intensifies due to the speed differences between vehicles on the main highway and those entering from the ramp. While connected and automated vehicles (CAVs) offer a promising solution for improving safety and easing congestion, their widespread adoption is still some way off. This means we need strategies that can effectively manage traffic flow when both human-driven vehicles (HDVs) and CAVs share the road.
Addressing this critical need, researchers have developed a new framework called HCOMC, which stands for Hierarchical Cooperative On-Ramp Merging Control. This innovative system is designed specifically for mixed traffic environments on two-lane highways, aiming to make the merging process safer, smoother, and more efficient.
The HCOMC framework builds upon existing models for how cars follow each other (longitudinal car-following) and change lanes (lateral lane-changing). It adapts these models to account for the distinct behaviors of both human drivers and automated vehicles, incorporating factors like human reaction times and advanced cooperative adaptive cruise control for CAVs.
How HCOMC Works
The HCOMC framework is composed of three main components:
Hierarchical Cooperative Planning Model: This model uses a modified “virtual vehicle” concept to coordinate the movements of vehicles, particularly when a CAV is involved in the merging process. It plans both the forward (longitudinal) and sideways (lateral) movements to ensure smooth integration into traffic.
Discretionary Lane-Changing Decision Model: Recognizing that lane changes significantly impact traffic flow and safety, this part of the framework uses game theory. It models the interactions between a vehicle wanting to change lanes and the vehicles around it, helping to determine the optimal and safest lane-changing decisions. This is especially important for CAVs, which can use information from surrounding vehicles to make strategic choices.
Multi-Objective Optimization Model: To achieve the best overall outcome, HCOMC employs an advanced optimization algorithm called NSGA-II. This algorithm considers multiple objectives simultaneously: enhancing safety, improving traffic efficiency, and reducing fuel consumption. It identifies the best merging position, trajectory, and cooperation mode for vehicles involved in the merge.
Also Read:
- Smart Lane Changes: Automated Vehicles Learn to Adapt to Human Trust
- Advanced AI Improves Autonomous Vehicle Eco-Driving and Traffic Flow
Key Benefits and Findings
The performance of HCOMC was rigorously tested through simulations under various conditions, including different traffic densities and varying rates of CAV penetration. The results highlight several significant advantages:
Enhanced Safety: HCOMC significantly improves the safety of all vehicles involved, increasing the critical distance between vehicles to prevent collisions.
Improved Stability and Rapidity: The framework helps stabilize traffic flow and speeds up the merging process, reducing the time it takes for traffic to normalize after a merge.
Optimized Traffic Efficiency: By minimizing low-speed regions, HCOMC contributes to a more efficient flow of traffic, even in congested merging zones.
Economized Fuel Consumption: The coordinated movements planned by HCOMC lead to smoother driving patterns, which in turn helps reduce overall fuel consumption.
Compared to existing benchmarks like the First-In-First-Out (FIFO) model and other game theory models, HCOMC consistently demonstrated superior comprehensive performance across most metrics. While there might be minor trade-offs in isolated scenarios, the framework consistently ensures safe car-following and lane-changing, and significantly boosts traffic efficiency and merging rapidity.
This research marks a substantial step forward in developing sophisticated control strategies for the evolving landscape of mixed traffic. It underscores the profound impact that CAV technologies can have on alleviating congestion and reducing accident risks at highway bottlenecks. For more detailed information, you can refer to the full research paper here.
