CoSafe: Cooperative Runtime Safety Assurance for Mixed-Traffic Autonomous Driving
CoSafe: Cooperative Runtime Safety Assurance for Mixed-Traffic Autonomous Driving
This paper introduces CoSafe, a modular and ADS-agnostic framework designed to ensure runtime safety in mixed-traffic environments by leveraging vehicle-to-everything (V2X) communication. The system enables roadside units to monitor scene-level hazards and coordinate interventions among multiple vehicles, addressing limitations of ego-centric approaches. Evaluated on the InterDrive benchmark with seven state-of-the-art autonomous driving systems, CoSafe achieves an average 70.03% improvement in driving scores, reduces safety violations by 87.80%, and identifies unsafe vehicles with an F3 score of 0.930 while meeting real-time constraints.
The authors present CoSafe, a cooperative runtime safety assurance framework integrated with seven autonomous driving systems that demonstrates improved driving scores and reduced safety violations on the InterDrive benchmark under mixed-traffic conditions.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
F3 score
0.93
View original evidence
rate CoSafe with seven state-of-the-art ADSes, and our evaluation on the InterDrive benchmark demonstrates that CoSafe improves the driving score of AVs by 70.03% on average, reduces safety violations by 87.80%, and accurately identifies unsafe AVs with an F3 score of 0.930, while satisfying real-time constraints.
Open source S5
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
rate CoSafe with seven state-of-the-art ADSes, and our evaluation on the InterDrive benchmark demonstrates that CoSafe improves the driving score of AVs by 70.03% on average, reduces safety violations by 87.80%, and accurately identifies unsafe AVs with an F3 score of 0.930, while satisfying real-time constraints.
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org