Problem
Research questions and friction points this paper is trying to address.
corner case
scenario generation
autonomous driving
risk percentile
adversity control
How corner is a corner case? Percentile control for highway scenario generation
🤖 AI Summary
This study addresses the lack of precise extremity control and the difficulty in quantifying "corner-case" severity in autonomous driving scenario generation. By defining scene risk as a percentile within a conditional distribution, it introduces a history-conditioned risk percentile request mechanism that unifies context-relative risk specification, physical realization, and evaluation criteria. Methodologically, a reference risk distribution is constructed to map physical objectives, while a percentile-conditional joint diffusion model combined with risk guidance at sampling time enables calibratable extreme multi-agent trajectory generation. Evaluated on the highD dataset, 98.75% of risk requests are satisfied within a 0.05 tolerance, achieving an average error of merely 0.00673.
📝 Abstract
Generating corner-case scenarios with appropriate adversity in a simulation environment is critical for testing an autonomous vehicle (AV) software stack's safety performance before deployment. Existing autonomous-driving scenario generators can enforce specific behavior, adversity, or feasibility conditions, but they provide limited control over how extreme a generated scenario is relative to plausible futures in the same traffic context. This study represents the adversity of a generated scenario as its percentile in the conditional distribution of future risk given the observed history. This view supports calibrated answers to two questions: how"corner"a generated corner-case scenario is and how its"cornerness"can be fine-tuned. To this end, we formulate history-conditioned risk-percentile requests and learn a reference risk distribution that maps each requested percentile to a physical risk target. We then use a percentile-conditioned joint diffusion model with sampling-time risk guidance to generate multi-agent futures, together with a reference-based criterion for evaluating percentile realization. Experiments use the minimum post-encroachment time (PET) between the ego and its surrounding vehicles as the risk surrogate on highD. On the primary evaluation set, our method realizes 1,422 of 1,440 requests within a 0.05 percentile tolerance (98.75%), with mean percentile error 0.00673 and PET-target error 0.00991 seconds. The resulting interface connects context-relative risk specification, physical realization, and evaluation through a common risk scale. Project website and videos of generated scenarios are available at https://hhj233.github.io/CornerPercentile/.
Innovation
Methods, ideas, or system contributions that make the work stand out.
corner-case generation
percentile control
conditional diffusion model
risk-guided sampling
autonomous driving simulation
🔎 Similar Papers
No similar papers found.
