Diffusion Models for Safety Validation of Autonomous Driving Systems
This addresses the problem of costly and risky real-world testing for autonomous vehicles, though it appears incremental as it applies existing diffusion methods to a specific domain.
The authors tackled the challenge of safety validation for autonomous driving systems by training a denoising diffusion model to generate realistic failure cases from initial traffic states, achieving results that capture diverse potential failures in four-way intersection scenarios.
Safety validation of autonomous driving systems is extremely challenging due to the high risks and costs of real-world testing as well as the rarity and diversity of potential failures. To address these challenges, we train a denoising diffusion model to generate potential failure cases of an autonomous vehicle given any initial traffic state. Experiments on a four-way intersection problem show that in a variety of scenarios, the diffusion model can generate realistic failure samples while capturing a wide variety of potential failures. Our model does not require any external training dataset, can perform training and inference with modest computing resources, and does not assume any prior knowledge of the system under test, with applicability to safety validation for traffic intersections.