SaFeR: Safety-Critical Scenario Generation for Autonomous Driving Test via Feasibility-Constrained Token Resampling

TL;DR

Closed-loop experiments on the Waymo Open Motion Dataset and nuPlan demonstrate that SaFeR significantly outperforms state-of-the-art baselines, achieving a higher solution rate and superior kinematic realism while maintaining strong adversarial effectiveness.

Abstract

Safety-critical scenario generation is crucial for evaluating autonomous driving systems. However, existing approaches often struggle to balance three conflicting objectives: adversarial criticality, physical feasibility, and behavioral realism. To bridge this gap, we propose SaFeR: safety-critical scenario generation for autonomous driving test via feasibility-constrained token resampling. We first formulate traffic generation as a discrete next token prediction problem, employing a Transformer-based model as a realism prior to capture naturalistic driving distributions. To capture complex interactions while effectively mitigating attention noise, we propose a novel differential attention mechanism within the realism prior. Building on this prior, SaFeR implements a novel resampling strategy that induces adversarial behaviors within a high-probability trust region to maintain naturalism, while enforcing a feasibility constraint derived from the Largest Feasible Region (LFR). By approximating the LFR via offline reinforcement learning, SaFeR effectively prevents the generation of theoretically inevitable collisions. Closed-loop experiments on the Waymo Open Motion Dataset and nuPlan demonstrate that SaFeR significantly outperforms state-of-the-art baselines, achieving a higher solution rate and superior kinematic realism while maintaining strong adversarial effectiveness.

Figures (5)

• Figure 1: Overview of SaFeR. The strategy resamples adversarial tokens from distributions generated by realism prior, constrained by the Largest Feasible Region.
• Figure 2: Pipeline of SaFeR. The pipeline consists of two core components: (I) Realism Prior Modeling utilizes a Transformer-based NTP model to learn naturalistic motion distributions via motion tokenization and differential attention. (II) Safety-Critical Token Resampling synthesizes the scenario by optimizing for adversarial criticality within a realistic Trust Region, enforced by the Largest Feasible Region constraint to enhance feasibility.
• Figure 3: Visualization of the learned LFR of ego vehicle. The red area depicts the Infeasible Region ($V_h^* > 0$), representing the set of spatial configurations for the critical background vehicle that lead to unavoidable collisions given the specified velocity vectors. The green area denotes the Largest Feasible Region (LFR) ($V_h^* \le 0$), where a feasible evasion policy theoretically exists for the ego vehicle.
• Figure 4: Qualitative comparison between the raw scenario and safety-critical scenario generated by SaFeR. The critical background vehicle is controlled by SaFeR, while the ego vehicle is controlled by DiffusionPlannerzheng2025difplan.
• Figure 5: The mean of Feasible Value Network $V_h$ during the training with multiple combinations of penalty constant $M$ and safety distance threshold $d_{th}$.