发表机构
School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院电气工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出图引导的安全扩散器（G2SD）框架，用高层拓扑图规划器引导低层扩散模型，解决扩散规划器的流形破裂问题，在Maze2D导航等任务中大幅提升无碰撞目标到达率与任务性能。
AI 中文摘要
许多基于扩散的规划器通过推理时引导来确保安全性，但这种交错的轨迹变形常因流形破裂而降低运动学可行性。我们提出图引导的安全扩散器（G2SD），这是一种利用高层拓扑图规划器引导低层扩散模型的分层框架。G2SD通过将数据流形抽象为学习到的潜在图并在其上执行高层规划，在结构层面确保安全性。扩散规划器生成连续轨迹，这些轨迹以高层规划器选择的图节点表示为条件。理论分析揭示了扩散规划器发生流形破裂的条件，并表明随着段数增加，G2SD通过降低约束违反概率提升安全性。实验表明，G2SD显著优于基线方法，在Maze2D导航中，无碰撞的目标到达率从40-50%提升至98%，且在运动任务中也取得了更优的任务得分。
英文摘要
Many diffusion-based planners enforce safety through inference-time guidance, but such interleaved trajectory deformations often degrade kinematic feasibility due to manifold rupture. We propose Graph-Guided Safe Diffuser (G2SD), a hierarchical framework that leverages a high-level topological graph planner to guide a low-level diffusion model. G2SD enforces safety at a structural level by abstracting the data manifold into a learned latent graph, on which high-level planning is performed. Continuous trajectories are generated by diffusion planners, which are conditioned on the graph node representations selected by the high-level planner. Theoretical analyses demonstrate conditions under which manifold rupture occurs in diffusion planners, and show that G2SD improves safety by reducing the constraint violation probability as the number of segments increases. Experiments demonstrate that G2SD substantially outperforms baselines, increasing goal-reaching rate without any collision from 40-50% to 98% in Maze2D navigation and also achieving superior task scores in locomotion.
Comments18 pages, 2 figures
