CEI: A Unified Interface for Cross-Embodiment Visuomotor Policy Learning in 3D Space
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917203288260608 |
|---|---|
| author | Wu, Tong Li, Shoujie Gong, Junhao Guo, Changqing Li, Xingting Mu, Shilong Ding, Wenbo |
| author_facet | Wu, Tong Li, Shoujie Gong, Junhao Guo, Changqing Li, Xingting Mu, Shilong Ding, Wenbo |
| contents | Robotic foundation models trained on large-scale manipulation datasets have shown promise in learning generalist policies, but they often overfit to specific viewpoints, robot arms, and especially parallel-jaw grippers due to dataset biases. To address this limitation, we propose Cross-Embodiment Interface (\CEI), a framework for cross-embodiment learning that enables the transfer of demonstrations across different robot arm and end-effector morphologies. \CEI introduces the concept of \textit{functional similarity}, which is quantified using Directional Chamfer Distance. Then it aligns robot trajectories through gradient-based optimization, followed by synthesizing observations and actions for unseen robot arms and end-effectors. In experiments, \CEI transfers data and policies from a Franka Panda robot to \textbf{16} different embodiments across \textbf{3} tasks in simulation, and supports bidirectional transfer between a UR5+AG95 gripper robot and a UR5+Xhand robot across \textbf{6} real-world tasks, achieving an average transfer ratio of 82.4\%. Finally, we demonstrate that \CEI can also be extended with spatial generalization and multimodal motion generation capabilities using our proposed techniques. Project website: https://cross-embodiment-interface.github.io/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_09163 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CEI: A Unified Interface for Cross-Embodiment Visuomotor Policy Learning in 3D Space Wu, Tong Li, Shoujie Gong, Junhao Guo, Changqing Li, Xingting Mu, Shilong Ding, Wenbo Robotics Robotic foundation models trained on large-scale manipulation datasets have shown promise in learning generalist policies, but they often overfit to specific viewpoints, robot arms, and especially parallel-jaw grippers due to dataset biases. To address this limitation, we propose Cross-Embodiment Interface (\CEI), a framework for cross-embodiment learning that enables the transfer of demonstrations across different robot arm and end-effector morphologies. \CEI introduces the concept of \textit{functional similarity}, which is quantified using Directional Chamfer Distance. Then it aligns robot trajectories through gradient-based optimization, followed by synthesizing observations and actions for unseen robot arms and end-effectors. In experiments, \CEI transfers data and policies from a Franka Panda robot to \textbf{16} different embodiments across \textbf{3} tasks in simulation, and supports bidirectional transfer between a UR5+AG95 gripper robot and a UR5+Xhand robot across \textbf{6} real-world tasks, achieving an average transfer ratio of 82.4\%. Finally, we demonstrate that \CEI can also be extended with spatial generalization and multimodal motion generation capabilities using our proposed techniques. Project website: https://cross-embodiment-interface.github.io/ |
| title | CEI: A Unified Interface for Cross-Embodiment Visuomotor Policy Learning in 3D Space |
| topic | Robotics |
| url | https://arxiv.org/abs/2601.09163 |