Learning Responsibility-Attributed Adversarial Scenarios for Testing Autonomous Vehicles

2026-05-13

AI Summary

This work addresses a critical limitation in existing adversarial simulation methods for autonomous driving: while effective at inducing collisions, they often fail to distinguish between genuine system failures and unavoidable traffic conflicts. To resolve this, the paper introduces CARS, a novel framework that integrates normative responsibility attribution into adversarial scenario generation. By incorporating context-aware adversary selection, optimized adversarial policy generation, and closed-loop simulation, CARS produces physically plausible collision scenarios that are demonstrably attributable to the autonomous system’s shortcomings. The approach embeds a “reasonably competent and careful driver” model derived from traffic regulations across multiple jurisdictions, enabling consistent generation of high-attributability scenarios on heterogeneous traffic datasets. This advances adversarial testing beyond mere collision discovery toward interpretable, regulation-compliant safety validation.

Abstract

Establishing trustworthy safety assurance for autonomous driving systems (ADSs) requires evidence that failures arise from avoidable system deficiencies rather than unavoidable traffic conflicts. Current adversarial simulation methods can efficiently expose collisions, but generally lack mechanisms to distinguish these fundamentally different failure modes. Here we present CARS (Context-Aware, Responsibility-attributed Scenario generation), a framework that integrates responsibility attribution directly into adversarial scenario generation. CARS combines context-aware adversary selection with a generative adversarial policy optimized in closed-loop simulation to construct collision scenarios that are both physically feasible and diagnostically attributable. Across benchmark datasets spanning heterogeneous national traffic environments, CARS consistently discovers feasible collision scenarios with high attribution rates under multiple regulation-prescribed careful and competent driver models. By coupling adversarial generation with normative responsibility assessment, CARS moves simulation testing beyond collision discovery toward the construction of interpretable, regulation-aligned safety evidence for scalable ADS validation.

Problem

• autonomous vehicles
• adversarial scenarios
• responsibility attribution
• safety validation
• collision scenarios

Innovation

• responsibility attribution
• adversarial scenario generation
• autonomous vehicle testing
• context-aware simulation
• safety validation