English

SEAL: Towards Safe Autonomous Driving via Skill-Enabled Adversary Learning for Closed-Loop Scenario Generation

Robotics 2025-07-15 v3 Artificial Intelligence Machine Learning

Abstract

Verification and validation of autonomous driving (AD) systems and components is of increasing importance, as such technology increases in real-world prevalence. Safety-critical scenario generation is a key approach to robustify AD policies through closed-loop training. However, existing approaches for scenario generation rely on simplistic objectives, resulting in overly-aggressive or non-reactive adversarial behaviors. To generate diverse adversarial yet realistic scenarios, we propose SEAL, a scenario perturbation approach which leverages learned objective functions and adversarial, human-like skills. SEAL-perturbed scenarios are more realistic than SOTA baselines, leading to improved ego task success across real-world, in-distribution, and out-of-distribution scenarios, of more than 20%. To facilitate future research, we release our code and tools: https://github.com/cmubig/SEAL

Keywords

Cite

@article{arxiv.2409.10320,
  title  = {SEAL: Towards Safe Autonomous Driving via Skill-Enabled Adversary Learning for Closed-Loop Scenario Generation},
  author = {Benjamin Stoler and Ingrid Navarro and Jonathan Francis and Jean Oh},
  journal= {arXiv preprint arXiv:2409.10320},
  year   = {2025}
}

Comments

Accepted to the IEEE Robotics and Automation Letters (RA-L) on June 28, 2025

R2 v1 2026-06-28T18:46:11.350Z