Beyond Task and Motion Planning: Hierarchical Robot Planning with General-Purpose Skills

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hedegaard, Benned, Wei, Yichen, Jaafar, Ahmed, Tellex, Stefanie, Konidaris, George, Shah, Naman
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917206933110784
author Hedegaard, Benned
Wei, Yichen
Jaafar, Ahmed
Tellex, Stefanie
Konidaris, George
Shah, Naman
author_facet Hedegaard, Benned
Wei, Yichen
Jaafar, Ahmed
Tellex, Stefanie
Konidaris, George
Shah, Naman
contents Task and motion planning is a well-established approach for solving long-horizon robot planning problems. However, traditional methods assume that each task-level robot action, or skill, can be reduced to kinematic motion planning. We address the challenge of combining motion planning with closed-loop motor controllers that go beyond mere kinematic considerations. We propose a novel framework that integrates these policies into motion planning using Composable Interaction Primitives (CIPs), enabling the use of diverse, non-composable pre-learned skills in hierarchical robot planning. We validate our Task and Skill Planning (TASP) approach through real-world experiments on a bimanual manipulator and a mobile manipulator, demonstrating that CIPs allow diverse robots to combine motion planning with general-purpose skills to solve complex, long-horizon tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2504_17901
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Beyond Task and Motion Planning: Hierarchical Robot Planning with General-Purpose Skills
Hedegaard, Benned
Wei, Yichen
Jaafar, Ahmed
Tellex, Stefanie
Konidaris, George
Shah, Naman
Robotics
Artificial Intelligence
Task and motion planning is a well-established approach for solving long-horizon robot planning problems. However, traditional methods assume that each task-level robot action, or skill, can be reduced to kinematic motion planning. We address the challenge of combining motion planning with closed-loop motor controllers that go beyond mere kinematic considerations. We propose a novel framework that integrates these policies into motion planning using Composable Interaction Primitives (CIPs), enabling the use of diverse, non-composable pre-learned skills in hierarchical robot planning. We validate our Task and Skill Planning (TASP) approach through real-world experiments on a bimanual manipulator and a mobile manipulator, demonstrating that CIPs allow diverse robots to combine motion planning with general-purpose skills to solve complex, long-horizon tasks.
title Beyond Task and Motion Planning: Hierarchical Robot Planning with General-Purpose Skills
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2504.17901