NOD-TAMP：基于神经对象描述符的可泛化长时序规划
NOD-TAMP: Generalizable Long-Horizon Planning with Neural Object Descriptors
AI总结：
针对长时序操作任务泛化难问题，提出结合神经对象描述符（NOD）与任务运动规划（TAMP）的NOD-TAMP框架，从少量演示提取轨迹并适配组合，在基准和真实任务中表现更优。
摘要：
在家庭和工厂场景中完成复杂操作任务仍具挑战性，原因在于长时序推理、细粒度交互以及广泛的物体与场景多样性。从演示中学习技能是一种有效策略，但这类方法在训练数据之外的泛化能力往往有限，且难以解决长时序任务。为克服这一问题，我们提出协同结合两种范式：可生成泛化性以物体为中心特征的神经对象描述符（Neural Object Descriptors, NODs），以及将短时序技能串联以解决多步任务的任务与运动规划（Task and Motion Planning, TAMP）框架。我们提出NOD-TAMP，这是一种基于TAMP的框架，可从少量人类演示中提取短操作轨迹，利用NOD特征调整这些轨迹，并将其组合以解决各类长时序、接触密集型任务。NOD-TAMP仅需少量演示即可解决现有操作基准测试，在需要多样泛化能力的新型桌面操作任务上，性能显著优于此前基于NOD的方法。最后，我们在多项真实世界任务中部署了NOD-TAMP，包括工具使用和高精度插入任务。更多详情请访问https://nodtamp.github.io/。
英文摘要：
Solving complex manipulation tasks in household and factory settings remains challenging due to long-horizon reasoning, fine-grained interactions, and broad object and scene diversity. Learning skills from demonstrations can be an effective strategy, but such methods often have limited generalizability beyond training data and struggle to solve long-horizon tasks. To overcome this, we propose to synergistically combine two paradigms: Neural Object Descriptors (NODs) that produce generalizable object-centric features and Task and Motion Planning (TAMP) frameworks that chain short-horizon skills to solve multi-step tasks. We introduce NOD-TAMP, a TAMP-based framework that extracts short manipulation trajectories from a handful of human demonstrations, adapts these trajectories using NOD features, and composes them to solve broad long-horizon, contact-rich tasks. NOD-TAMP solves existing manipulation benchmarks with a handful of demonstrations and significantly outperforms prior NOD-based approaches on new tabletop manipulation tasks that require diverse generalization. Finally, we deploy NOD-TAMP on a number of real-world tasks, including tool-use and high-precision insertion. For more details, please visit https://nodtamp.github.io/.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
