用检索增强生成技术，让自动驾驶规划更安全、可控。
RealDrive: Retrieval-Augmented Driving with Diffusion Models
- 从数据中检索相似驾驶场景，用扩散模型融合生成轨迹。
- 在Waymo数据集上碰撞率降低40%，长尾场景泛化能力更强。
- 基于任务的检索器比通用检索器更适合自动驾驶规划。
基于学习的规划器通过从数据中学习复杂交互，生成类人驾驶行为，克服了规则驱动规划器僵硬的问题。然而，数据驱动方法在罕见且高危场景中表现不佳，且生成轨迹的可控性有限。为此，我们提出RealDrive，一种基于检索增强生成（RAG）的框架，通过从训练数据集中检索最相关的专家示范来初始化基于扩散模型的规划策略。通过在当前观测与检索样本间进行去噪插值，该方法实现了细粒度控制与多样化安全行为，利用检索到场景的强大先验知识。另一个关键发现是：使用规划目标训练的任务相关检索器，在本框架中相比任务无关检索器带来更优的规划性能。实验表明，相较于标准学习型规划器，RealDrive在长尾事件上具备更强泛化能力，并提升轨迹多样性；在Waymo Open Motion数据集上，碰撞率降低了40%。
原文摘要 · Abstract (English)
Learning-based planners generate natural human-like driving behaviors by learning to reason about nuanced interactions from data, overcoming the rigid behaviors that arise from rule-based planners. Nonetheless, data-driven approaches often struggle with rare, safety-critical scenarios and offer limited controllability over the generated trajectories. To address these challenges, we propose RealDrive, a Retrieval-Augmented Generation (RAG) framework that initializes a diffusion-based planning policy by retrieving the most relevant expert demonstrations from the training dataset. By interpolating between current observations and retrieved examples through a denoising process, our approach enables fine-grained control and safe behavior across diverse scenarios, leveraging the strong prior provided by the retrieved scenario. Another key insight we produce is that a task-relevant retrieval model trained with planning-based objectives results in superior planning performance in our framework compared to a task-agnostic retriever. Experimental results demonstrate improved generalization to long-tail events and enhanced trajectory diversity compared to standard learning-based planners -- we observe a 40% reduction in collision rate on the Waymo Open Motion dataset with RAG.
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。