H$^3$DP: Triply-Hierarchical Diffusion Policy for Visuomotor Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lu, Yiyang, Tian, Yufeng, Yuan, Zhecheng, Wang, Xianbang, Hua, Pu, Xue, Zhengrong, Xu, Huazhe
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912435129024512
author Lu, Yiyang
Tian, Yufeng
Yuan, Zhecheng
Wang, Xianbang
Hua, Pu
Xue, Zhengrong
Xu, Huazhe
author_facet Lu, Yiyang
Tian, Yufeng
Yuan, Zhecheng
Wang, Xianbang
Hua, Pu
Xue, Zhengrong
Xu, Huazhe
contents Visuomotor policy learning has witnessed substantial progress in robotic manipulation, with recent approaches predominantly relying on generative models to model the action distribution. However, these methods often overlook the critical coupling between visual perception and action prediction. In this work, we introduce $\textbf{Triply-Hierarchical Diffusion Policy}~(\textbf{H$^{\mathbf{3}}$DP})$, a novel visuomotor learning framework that explicitly incorporates hierarchical structures to strengthen the integration between visual features and action generation. H$^{3}$DP contains $\mathbf{3}$ levels of hierarchy: (1) depth-aware input layering that organizes RGB-D observations based on depth information; (2) multi-scale visual representations that encode semantic features at varying levels of granularity; and (3) a hierarchically conditioned diffusion process that aligns the generation of coarse-to-fine actions with corresponding visual features. Extensive experiments demonstrate that H$^{3}$DP yields a $\mathbf{+27.5\%}$ average relative improvement over baselines across $\mathbf{44}$ simulation tasks and achieves superior performance in $\mathbf{4}$ challenging bimanual real-world manipulation tasks. Project Page: https://lyy-iiis.github.io/h3dp/.
format Preprint
id arxiv_https___arxiv_org_abs_2505_07819
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle H$^3$DP: Triply-Hierarchical Diffusion Policy for Visuomotor Learning
Lu, Yiyang
Tian, Yufeng
Yuan, Zhecheng
Wang, Xianbang
Hua, Pu
Xue, Zhengrong
Xu, Huazhe
Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
Visuomotor policy learning has witnessed substantial progress in robotic manipulation, with recent approaches predominantly relying on generative models to model the action distribution. However, these methods often overlook the critical coupling between visual perception and action prediction. In this work, we introduce $\textbf{Triply-Hierarchical Diffusion Policy}~(\textbf{H$^{\mathbf{3}}$DP})$, a novel visuomotor learning framework that explicitly incorporates hierarchical structures to strengthen the integration between visual features and action generation. H$^{3}$DP contains $\mathbf{3}$ levels of hierarchy: (1) depth-aware input layering that organizes RGB-D observations based on depth information; (2) multi-scale visual representations that encode semantic features at varying levels of granularity; and (3) a hierarchically conditioned diffusion process that aligns the generation of coarse-to-fine actions with corresponding visual features. Extensive experiments demonstrate that H$^{3}$DP yields a $\mathbf{+27.5\%}$ average relative improvement over baselines across $\mathbf{44}$ simulation tasks and achieves superior performance in $\mathbf{4}$ challenging bimanual real-world manipulation tasks. Project Page: https://lyy-iiis.github.io/h3dp/.
title H$^3$DP: Triply-Hierarchical Diffusion Policy for Visuomotor Learning
topic Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.07819