用大模型在线生成危险驾驶场景，提升自动驾驶测试效率
Seeking to Collide: Online Safety-Critical Scenario Generation for Autonomous Driving with Retrieval Augmented Large Language Models
- 基于大模型分析车辆意图，实时生成对抗性行驶轨迹
- 最小碰撞时间从1.62秒降至1.08秒，碰撞率达75%
- 动态记忆库自动扩展行为库，适合安全测试人员使用
基于仿真的测试对验证自动驾驶车辆至关重要，但现有场景生成方法要么过度拟合常见驾驶模式，要么以离线非交互方式运行，无法暴露罕见的安全关键边缘案例。本文提出一种在线、检索增强的大语言模型（LLM）框架，用于生成安全关键驾驶场景。该方法首先利用基于LLM的行为分析器，从观测状态中推断背景车辆的最危险意图，随后调用额外的LLM代理合成可行的对抗性轨迹。为缓解灾难性遗忘并加速适应，系统引入动态记忆与检索库，自动在发现新意图时扩展意图-规划器对库。在Waymo Open Motion Dataset上的评估表明，该模型将平均最小碰撞时间从1.62秒降低至1.08秒，碰撞率高达75%，显著优于基线方法。
原文摘要 · Abstract (English)
Simulation-based testing is crucial for validating autonomous vehicles (AVs), yet existing scenario generation methods either overfit to common driving patterns or operate in an offline, non-interactive manner that fails to expose rare, safety-critical corner cases. In this paper, we introduce an online, retrieval-augmented large language model (LLM) framework for generating safety-critical driving scenarios. Our method first employs an LLM-based behavior analyzer to infer the most dangerous intent of the background vehicle from the observed state, then queries additional LLM agents to synthesize feasible adversarial trajectories. To mitigate catastrophic forgetting and accelerate adaptation, we augment the framework with a dynamic memorization and retrieval bank of intent-planner pairs, automatically expanding its behavioral library when novel intents arise. Evaluations using the Waymo Open Motion Dataset demonstrate that our model reduces the mean minimum time-to-collision from 1.62 to 1.08 s and incurs a 75% collision rate, substantially outperforming baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。