Sequencing-enabled Hierarchical Cooperative CAV On-ramp Merging Control with Enhanced Stability and Feasibility

Sixu Li, Yang Zhou, Xinyue Ye, Jiwan Jiang, Meng Wang

Abstract

This paper develops a sequencing-enabled hierarchical connected automated vehicle (CAV) cooperative on-ramp merging control framework. The proposed framework consists of a two-layer design: the upper level control sequences the vehicles to harmonize the traffic density across mainline and on-ramp segments while enhancing lower-level control efficiency through a mixed-integer linear programming formulation. Subsequently, the lower-level control employs a longitudinal distributed model predictive control (MPC) supplemented by a virtual car-following (CF) concept to ensure asymptotic local stability, $l_2$ norm string stability, and safety. Proofs of asymptotic local stability and $l_2$ norm string stability are mathematically derived. Compared to other prevalent asymptotic local-stable MPC controllers, the proposed distributed MPC controller greatly expands the initial feasible set. Additionally, an auxiliary lateral control is developed to maintain lane-keeping and merging smoothness while accommodating ramp geometric curvature. To validate the proposed framework, multiple numerical experiments are conducted. Results indicate a notable outperformance of our upper-level controller against a distance-based sequencing method. Furthermore, the lower-level control effectively ensures smooth acceleration, safe merging with adequate spacing, adherence to proven longitudinal local and string stability, and rapid regulation of lateral deviations.

Introduction

On-ramp merging is a critical bottleneck in modern transportation, often acting as a "traffic void" that triggers congestion and increases safety risks. While macroscopic tools like ramp metering exist, they often fail to account for the micro-level physics of individual vehicles. This paper introduces a hierarchical control framework for Connected Automated Vehicles (CAVs) designed to master the complexities of on-ramp merging. By integrating a high-level sequencing algorithm with a low-level distributed Model Predictive Control (MPC) system, the framework achieves guaranteed traffic stability and a significantly wider "feasible set," allowing the system to safely handle vehicles even when they enter the control area far from their ideal merging positions.

Methodology

The authors propose a two-layer hierarchical design to manage the merging process efficiently.

Upper-Level Sequencing

The upper level uses Mixed-Integer Linear Programming (MILP) to determine the optimal merging sequence. Unlike standard First-In-First-Out (FIFO) methods that only consider distance, this approach balances traffic density between the mainline and the ramp to maximize overall throughput and reduce convergence time.

Lower-Level Control

The lower level manages the actual driving dynamics through two components:

  1. Longitudinal Control: This employs a Distributed MPC (DMPC). A key innovation is treating merging as a "virtual car-following" problem. Vehicles on the mainline and the on-ramp are projected onto a single Virtual Z-axis, effectively transforming a complex merging scenario into a standard platoon management task.
The virtual Z-axis concept maps vehicles from the mainline and on-ramp onto a single virtual coordinate system for simplified control.
  1. Lateral Control: A linearized MPC based on a kinematic bicycle model ensures vehicles maintain lane centering, even through the sharp curvatures typical of on-ramps.

Stability and Feasibility

A major contribution of this work is the expansion of the initial feasible set. Traditional MPC controllers often require vehicles to start very close to an equilibrium state to guarantee stability. By carefully designing terminal constraints and safety cost functions, this framework allows for a much broader range of initial speeds and positions.

The paper provides rigorous mathematical proofs for:

  • Asymptotic Local Stability: The ability of a single vehicle to return to its equilibrium state after a disturbance.
  • $l_2$ Norm String Stability: The ability of a platoon to dampen disturbances as they propagate backward, preventing "phantom" traffic jams.
Space-time diagram illustrating the real, predicted, and assumed trajectories of CAVs within the MPC framework to ensure safe spacing and stability.

Results and Discussion

The framework was validated using numerical simulations, incorporating real-world acceleration profiles from the NGSIM (Next Generation Simulation) US Highway 101 dataset to provide realistic disturbances.

  • Sequencing Efficiency: The MILP-based sequencing outperformed distance-based methods, leading to smoother driving and faster convergence to safe spacing.
  • String Stability: In tests with 10 CAVs, disturbances from the lead vehicle were successfully dampened, confirming the theoretical stability proofs.
  • Real-Time Performance: Computation times for the lower-level controller averaged between 0.045s and 0.047s, well within the 0.1s threshold required for real-time high-speed automated driving.

Conclusion

While the current framework focuses on single-lane scenarios and assumes a 100% CAV environment with perfect communication, it represents a significant step forward in cooperative merging. Future research will explore multi-lane maneuvers, mixed traffic conditions involving human-driven vehicles, and the impact of communication latencies on stability guarantees.

Summary updated

Abstract

This paper develops a sequencing-enabled hierarchical connected automated vehicle (CAV) cooperative on-ramp merging control framework. The proposed framework consists of a two-layer design: the upper level control sequences the vehicles to harmonize the traffic density across mainline and on-ramp segments while enhancing lower-level control efficiency through a mixed-integer linear programming formulation. Subsequently, the lower-level control employs a longitudinal distributed model predictive control (MPC) supplemented by a virtual car-following (CF) concept to ensure asymptotic local stability, l_2 norm string stability, and safety. Proofs of asymptotic local stability and l_2 norm string stability are mathematically derived. Compared to other prevalent asymptotic local-stable MPC controllers, the proposed distributed MPC controller greatly expands the initial feasible set. Additionally, an auxiliary lateral control is developed to maintain lane-keeping and merging smoothness while accommodating ramp geometric curvature. To validate the proposed framework, multiple numerical experiments are conducted. Results indicate a notable outperformance of our upper-level controller against a distance-based sequencing method. Furthermore, the lower-level control effectively ensures smooth acceleration, safe merging with adequate spacing, adherence to proven longitudinal local and string stability, and rapid regulation of lateral deviations.
BibTeX
@article{li2024sequencing,
  author = {Sixu Li and Yang Zhou and Xinyue Ye and Jiwan Jiang and Meng Wang},
  title = {{Sequencing-enabled Hierarchical Cooperative CAV On-ramp Merging Control with Enhanced Stability and Feasibility}},
  journal = {IEEE Transactions on Intelligent Vehicles},
  year = {2024},
  doi = {10.1109/tiv.2024.3409381},
}
Figure