# KineWorld：用于具身世界建模的动作诱导传输场

> 原文链接：https://robotics.hellobike.com/papers/2026-10-05-2610-06349
> 频道：papers | 更新日期：2026-10-05 | 标签：arxiv, cs.RO
> 核心摘要：KineWorld用机器人运动构建传输场并重加权生成监督，在双臂操作数据上验证动作后果预测。

# KineWorld：用于具身世界建模的动作诱导传输场

::: v-pre
> 原文链接：[KineWorld: Action-Induced Transport Fields for Embodied World Modeling](https://arxiv.org/abs/2610.06349)（arXiv 2610.06349 · 发布日期：2026-10-05）

**KineWorld: Action-Induced Transport Fields for Embodied World Modeling**

KineWorld用机器人运动构建传输场并重加权生成监督，在双臂操作数据上验证动作后果预测。

## 核心要点

- 现有动作条件世界模型常用均匀视觉生成目标，可能低估交互中关键的稀疏空间变化。
- KTL从指令机器人运动构建渲染器导出、相机对齐的传输场，用于分配生成监督。
- TAWD在视频潜变量网格上校准运动支持，并以均匀与传输聚焦分布的混合重加权未来RGB流匹配。
- 在RoboTwin 2.0的ALOHA-AgileX双臂操作数据上训练，单视角EWMScore-P为68.95，多视角TWB-Score为54.82。

**验证范围与局限**：验证范围为RoboTwin 2.0中ALOHA-AgileX双臂操作数据的单视角和多视角评估。

## 英文摘要 (Abstract)

Embodied world models predict the visual consequences of candidate actions before execution. However, existing action-conditioned world models often adopt uniformly weighted visual generation objectives that can be misaligned with embodied prediction needs. Even with explicit motion conditioning, these objectives can underemphasize spatially sparse changes that are critical to interaction. We propose KineWorld, a transport-aware world-modeling framework that extends robot kinematics from motion conditioning to the spatial allocation of generative supervision. Kinematic Transport Lifting (KTL) constructs renderer-derived, camera-aligned transport fields from commanded robot motion. Transport-Aware World Diffusion (TAWD) calibrates their motion support on the video-latent grid and reweights future-RGB flow matching through a normalized mixture of uniform and transport-focused distributions. We train KineWorld using ALOHA-AgileX bimanual manipulation data from RoboTwin 2.0. KineWorld achieves an EWMScore-P of 68.95 in single-view evaluation and a TWB-Score of 54.82 in multi-view evaluation. These results support a shift from appearance fitting toward action-consequence modeling for embodied decision-making.

作者：Ziying Song, Yuchen Liu, Zhuoran Xu, Ziyang Liu, Jian Jin, Jiangtao Su, Haibao Yu, Lei Yang, Yuanpei Chen
:::

## 原始参考来源 (Sources)

- [arXiv 2610.06349：KineWorld: Action-Induced Transport Fields for Embodied World Modeling](https://arxiv.org/abs/2610.06349)