Learning Semantic Atomic Skills for Multi-Task Robotic Manipulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Yihang, Wang, Weiqing, Wu, Shijie, Shi, Ye, Wang, Jingya
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909972467548160
author Zhu, Yihang
Wang, Weiqing
Wu, Shijie
Shi, Ye
Wang, Jingya
author_facet Zhu, Yihang
Wang, Weiqing
Wu, Shijie
Shi, Ye
Wang, Jingya
contents While imitation learning has shown impressive results in single-task robot manipulation, scaling it to multi-task settings remains a fundamental challenge due to issues such as suboptimal demonstrations, trajectory noise, and behavioral multi-modality. Existing skill-based methods attempt to address this by decomposing actions into reusable abstractions, but they often rely on fixed-length segmentation or environmental priors that limit semantic consistency and cross-task generalization. In this work, we propose AtomSkill, a novel multi-task imitation learning framework that learns and leverages a structured Atomic Skill Space for composable robot manipulation. Our approach is built on two key technical contributions. First, we construct a Semantically Grounded Atomic Skill Library by partitioning demonstrations into variable-length skills using gripper-state keyframe detection and vision-language model annotation. A contrastive learning objective ensures the resulting skill embeddings are both semantically consistent and temporally coherent. Second, we propose an Action Generation module with Keypose Imagination, which jointly predicts a skill's long-horizon terminal keypose and its immediate action sequence. This enables the policy to reason about overarching motion goals and fine-grained control simultaneously, facilitating robust skill chaining. Extensive experiments in simulated and real-world environments show that AtomSkill consistently outperforms state-of-the-art methods across diverse manipulation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2512_18368
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Semantic Atomic Skills for Multi-Task Robotic Manipulation
Zhu, Yihang
Wang, Weiqing
Wu, Shijie
Shi, Ye
Wang, Jingya
Robotics
While imitation learning has shown impressive results in single-task robot manipulation, scaling it to multi-task settings remains a fundamental challenge due to issues such as suboptimal demonstrations, trajectory noise, and behavioral multi-modality. Existing skill-based methods attempt to address this by decomposing actions into reusable abstractions, but they often rely on fixed-length segmentation or environmental priors that limit semantic consistency and cross-task generalization. In this work, we propose AtomSkill, a novel multi-task imitation learning framework that learns and leverages a structured Atomic Skill Space for composable robot manipulation. Our approach is built on two key technical contributions. First, we construct a Semantically Grounded Atomic Skill Library by partitioning demonstrations into variable-length skills using gripper-state keyframe detection and vision-language model annotation. A contrastive learning objective ensures the resulting skill embeddings are both semantically consistent and temporally coherent. Second, we propose an Action Generation module with Keypose Imagination, which jointly predicts a skill's long-horizon terminal keypose and its immediate action sequence. This enables the policy to reason about overarching motion goals and fine-grained control simultaneously, facilitating robust skill chaining. Extensive experiments in simulated and real-world environments show that AtomSkill consistently outperforms state-of-the-art methods across diverse manipulation tasks.
title Learning Semantic Atomic Skills for Multi-Task Robotic Manipulation
topic Robotics
url https://arxiv.org/abs/2512.18368