Safety-Critical Traffic Control for Mixed Autonomy Systems with Input Delay and Disturbances

Zhao, Chenguang; Yu, Huan · 2023 · Unknown

DOI: 10.1109/cdc49753.2023.10383945

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 safety challenges in mixed-autonomy traffic systems, where connected automated vehicles (CAVs) coexist with human-driven vehicles (HVs). While existing controllers often focus on stabilizing traffic flow and reducing stop-and-go waves, they may fail to prevent rear-end collisions, particularly when subject to input delays and external disturbances. The authors propose a Robust Safety-Critical Traffic Control (RSTC) framework that guarantees safety for both the CAV and following HVs despite these uncertainties. The motivation stems from the practical reality that CAVs experience significant input delays due to communication, processing, and actuation lags, while the speed of leading HVs acts as an unpredictable disturbance. The methodology employs Control Barrier Functions (CBF) to enforce safety constraints on a nominal controller designed for string stability. To handle input delay, the authors design a state predictor that estimates the future system state over the delay interval using current states and historical inputs. Since the future speed of the leading HV is unavailable, the predictor assumes the current disturbance persists, introducing a prediction error. The authors derive bounds on this error under the assumption that the leading vehicle’s acceleration is bounded. They then construct robust CBF candidates with relative degree one for both the CAV and HVs. These constraints account for the prediction error and disturbance bounds, ensuring the forward invariance of a safe set defined by a Constant Time Headway spacing policy. The resulting safe control input is synthesized via a Quadratic Programming (QP) formulation that minimizes deviation from the nominal stabilizing controller while satisfying the robust safety constraints. The main theoretical finding is the proof that the proposed RSTC guarantees the forward invariance of the safe set for all vehicles in the platoon, provided the derivative of the speed disturbance (acceleration) is bounded. The authors demonstrate that the robust CBF design successfully compensates for the inaccuracies introduced by the delay-compensating predictor and external disturbances. Numerical simulations validate the approach in two safety-critical scenarios prone to rear-end collisions. The results show that the RSTC controller maintains safe spacing gaps and prevents collisions, whereas the nominal controller without safety constraints fails to do so under similar conditions. The significance of this work lies in its integration of delay compensation and disturbance robustness into safety-critical control for traffic systems. By providing a theoretical guarantee of safety under practical uncertainties like input delay and unpredictable human driver behavior, the RSTC framework offers a viable solution for deploying CAVs in mixed-autonomy environments. This contributes to the broader field by addressing a critical gap in the literature: ensuring that stabilizing traffic controllers do not compromise safety, thereby enhancing public acceptance and operational reliability of automated driving technologies.

Key finding

The proposed Robust Safety-critical Traffic Control framework guarantees the forward invariance of safe sets for mixed-autonomy traffic systems under input delay and bounded external disturbances, thereby preventing rear-end collisions in scenarios where nominal stabilizing controllers fail.

Methodology

simulation_modeling

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. Discovered via author_sweep_intake on 2026-05-28.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

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

• teleoperation remote driving