Spatial Policy: Guiding Visuomotor Robotic Manipulation with Spatial-Aware Modeling and Reasoning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Liu, Yijun, Liu, Yuwei, Meng, Yuan, Zhang, Jieheng, Zhou, Yuwei, Li, Ye, Jiang, Jiacheng, Ji, Kangye, Ge, Shijia, Wang, Zhi, Zhu, Wenwu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909910130753536
author Liu, Yijun
Liu, Yuwei
Meng, Yuan
Zhang, Jieheng
Zhou, Yuwei
Li, Ye
Jiang, Jiacheng
Ji, Kangye
Ge, Shijia
Wang, Zhi
Zhu, Wenwu
author_facet Liu, Yijun
Liu, Yuwei
Meng, Yuan
Zhang, Jieheng
Zhou, Yuwei
Li, Ye
Jiang, Jiacheng
Ji, Kangye
Ge, Shijia
Wang, Zhi
Zhu, Wenwu
contents Vision-centric hierarchical embodied models have demonstrated strong potential. However, existing methods lack spatial awareness capabilities, limiting their effectiveness in bridging visual plans to actionable control in complex environments. To address this problem, we propose Spatial Policy (SP), a unified spatial-aware visuomotor robotic manipulation framework via explicit spatial modeling and reasoning. Specifically, we first design a spatial-conditioned embodied video generation module to model spatially guided predictions through the spatial plan table. Then, we propose a flow-based action prediction module to infer executable actions with coordination. Finally, we propose a spatial reasoning feedback policy to refine the spatial plan table via dual-stage replanning. Extensive experiments show that SP substantially outperforms state-of-the-art baselines, achieving over 33% improvement on Meta-World and over 25% improvement on iTHOR, demonstrating strong effectiveness across 23 embodied control tasks. We additionally evaluate SP in real-world robotic experiments to verify its practical viability. SP enhances the practicality of embodied models for robotic control applications. Code and checkpoints are maintained at https://plantpotatoonmoon.github.io/SpatialPolicy/.
format Preprint
id arxiv_https___arxiv_org_abs_2508_15874
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Spatial Policy: Guiding Visuomotor Robotic Manipulation with Spatial-Aware Modeling and Reasoning
Liu, Yijun
Liu, Yuwei
Meng, Yuan
Zhang, Jieheng
Zhou, Yuwei
Li, Ye
Jiang, Jiacheng
Ji, Kangye
Ge, Shijia
Wang, Zhi
Zhu, Wenwu
Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
Vision-centric hierarchical embodied models have demonstrated strong potential. However, existing methods lack spatial awareness capabilities, limiting their effectiveness in bridging visual plans to actionable control in complex environments. To address this problem, we propose Spatial Policy (SP), a unified spatial-aware visuomotor robotic manipulation framework via explicit spatial modeling and reasoning. Specifically, we first design a spatial-conditioned embodied video generation module to model spatially guided predictions through the spatial plan table. Then, we propose a flow-based action prediction module to infer executable actions with coordination. Finally, we propose a spatial reasoning feedback policy to refine the spatial plan table via dual-stage replanning. Extensive experiments show that SP substantially outperforms state-of-the-art baselines, achieving over 33% improvement on Meta-World and over 25% improvement on iTHOR, demonstrating strong effectiveness across 23 embodied control tasks. We additionally evaluate SP in real-world robotic experiments to verify its practical viability. SP enhances the practicality of embodied models for robotic control applications. Code and checkpoints are maintained at https://plantpotatoonmoon.github.io/SpatialPolicy/.
title Spatial Policy: Guiding Visuomotor Robotic Manipulation with Spatial-Aware Modeling and Reasoning
topic Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.15874