发表机构
Carnegie Mellon University; National University of Singapore; DEVCOM Army Research Laboratory; Harvard University（卡内基梅隆大学; 新加坡国立大学; DEVCOM陆军研究实验室; 哈佛大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出流形引导的轨迹规划方法，仅用状态观测学习状态空间流形并利用其几何构造轨迹，实现对未见约束的泛化，在基准上表现优异且无需轨迹数据。
AI 中文摘要
轨迹规划的一种常见方式是利用在大量专家轨迹上训练生成模型。在推理时，模型通过以任务目标约束为条件生成可执行的轨迹。然而，基于轨迹的方法依赖昂贵的监督，随序列长度扩展性差，且通常对未见约束（如新颖的起点-目标点对）泛化能力不佳。我们提出一种替代方法，学习底层状态空间流形，并利用流形的几何结构进行轨迹规划。该方法仅需状态观测，并可通过在学习的状态空间流形上构造轨迹，实现对未见约束的泛化。在Maze2D和机器人运动规划基准上的实验表明，Ariadne能从仅状态监督中构造可行路径，并泛化到未见过的起点-目标组合。在高维双臂规划中，它仍能与轨迹监督和经典规划器保持竞争力，同时训练时无需任何轨迹数据。
英文摘要
A common way for trajectory planning is to leverage generative models trained on large collections of expert trajectories. At inference time, the model generates executable trajectories by conditioning on task goal constraints. However, trajectory-based methods rely on costly supervision, scale poorly with sequence length, and often generalize poorly to unseen constraints such as novel start-goal pairs. We propose an alternative to learn the underlying state-space manifold and use the geometry of the manifold for trajectory planning. This approach requires only state observations and enables generalization to unseen constraints by con- structing trajectories on the learned manifold of the state space. Experiments on Maze2D and robotic motion-planning benchmarks show that Ariadne constructs feasible paths from state-only supervision and generalizes to unseen start-goal combinations. On high-dimensional dual-arm planning, it remains competitive with trajectory-supervised and classical planners, while requiring no trajectory data for training.
CommentsAccepted at NeurIPS 2026. Code can be found in https://github.com/SilongYong/Ariadne/
