Safety-critical traffic control by connected automated vehicles

Zhao, Chenguang; Yu, Huan; Molnár, Tamás G. · 2023 · OpenAlex-citations

DOI: 10.1016/j.trc.2023.104230

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of stabilizing mixed-autonomy traffic flows while ensuring formal safety guarantees, a gap in existing research where traffic stabilization and safety control are typically treated separately. The authors propose Safety-Critical Traffic Control (STC), a strategy enabling a single Connected Automated Vehicle (CAV) to dampen traffic oscillations (achieving string stability) behind it while maintaining collision-free distances relative to both the preceding vehicle and following Connected Human-Driven Vehicles (HDVs). The motivation stems from the prevalence of stop-and-go congestion and the need for provable safety in transitional traffic environments where CAVs and HDVs coexist. The methodology employs a three-layer hierarchical control framework. The top layer defines desired steady states for spacing and velocity. The middle layer utilizes a nominal stabilizing controller, specifically Leading Cruise Control (LCC), designed to mitigate traffic oscillations based on linearized traffic dynamics and head-to-tail string stability criteria. The bottom layer implements a safety filter using Control Barrier Functions (CBFs) to modify the nominal control input. This filter solves a quadratic program that minimizes deviation from the nominal controller while satisfying CBF-based safety constraints. To handle scenarios where the CAV lacks full state information of following HDVs, the authors integrate state observers (specifically Luenberger observers) to estimate unmeasured states, extending the framework to output feedback. Safety is enforced through policies such as Time Headway, Time-to-Collision, and Stopping Distance Headway, with hard constraints applied to the CAV and soft constraints to following HDVs. The study validates the STC framework through extensive numerical simulations incorporating real-world vehicle trajectory data for the head vehicle. The results demonstrate that the proposed approach successfully achieves string stability, effectively attenuating velocity fluctuations along the vehicle chain, while simultaneously providing provable safety guarantees. The simulations confirm that the CAV maintains safe distances from the preceding vehicle regardless of its behavior and ensures safe spacing for following HDVs, even when state information is partially unavailable and estimated via observers. The analysis also identifies the limitations of the approach and analyzes the impact of various control parameters on performance. The significance of this work lies in its integration of traffic stabilization and safety-critical control into a unified, provably safe framework. By leveraging CBFs, the authors provide formal guarantees that the CAV will not collide with neighbors while still performing its role in smoothing traffic flow. This addresses a critical need in mixed-autonomy systems, offering a robust solution that balances efficiency and safety. The use of state observers further enhances practical applicability by accommodating realistic connectivity limitations. The findings suggest that single CAVs can significantly improve traffic conditions in mixed fleets without compromising safety, providing a theoretical and practical foundation for future deployment of connected automated vehicles.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• following distance