NOD-TAMP: Generalizable Long-Horizon Planning with Neural Object Descriptors

Shuo Cheng, Caelan Reed Garrett, Ajay Mandlekar, Danfei Xu

Keywords: Robot Learning, Robot Planning, Manipulation

TL;DR: We introduce NOD-TAMP, a TAMP-based framework that can solve broad long-horizon manipulation tasks by adapting and composing short manipulation trajectories from a handful of human demonstration.

Abstract: 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/.

Website: https://nodtamp.github.io/

Student Paper: yes

Submission Number: 434