SEAL: Towards Safe Autonomous Driving via Skill-Enabled Adversary Learning for Closed-Loop Scenario Generation
This work addresses safety-critical scenario generation for autonomous driving systems, offering an incremental improvement over existing methods by enhancing realism and effectiveness in testing.
The paper tackled the problem of generating realistic adversarial scenarios for autonomous driving verification by proposing SEAL, which uses learned objectives and human-like skills to create more realistic perturbations, resulting in over 20% improvement in ego task success across various scenarios.
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