U-DiT Policy: U-shaped Diffusion Transformers for Robotic Manipulation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908565588934656 |
|---|---|
| author | Wu, Linzhi Mei, Aoran Wang, Xiyue Zhu, Guo-Niu Gan, Zhongxue |
| author_facet | Wu, Linzhi Mei, Aoran Wang, Xiyue Zhu, Guo-Niu Gan, Zhongxue |
| contents | Diffusion-based methods have been acknowledged as a powerful paradigm for end-to-end visuomotor control in robotics. Most existing approaches adopt a Diffusion Policy in U-Net architecture (DP-U), which, while effective, suffers from limited global context modeling and over-smoothing artifacts. To address these issues, we propose U-DiT Policy, a novel U-shaped Diffusion Transformer framework. U-DiT preserves the multi-scale feature fusion advantages of U-Net while integrating the global context modeling capability of Transformers, thereby enhancing representational power and policy expressiveness. We evaluate U-DiT extensively across both simulation and real-world robotic manipulation tasks. In simulation, U-DiT achieves an average performance gain of 10\% over baseline methods and surpasses Transformer-based diffusion policies (DP-T) that use AdaLN blocks by 6\% under comparable parameter budgets. On real-world robotic tasks, U-DiT demonstrates superior generalization and robustness, achieving an average improvement of 22.5\% over DP-U. In addition, robustness and generalization experiments under distractor and lighting variations further highlight the advantages of U-DiT. These results highlight the effectiveness and practical potential of U-DiT Policy as a new foundation for diffusion-based robotic manipulation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_24579 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | U-DiT Policy: U-shaped Diffusion Transformers for Robotic Manipulation Wu, Linzhi Mei, Aoran Wang, Xiyue Zhu, Guo-Niu Gan, Zhongxue Robotics Diffusion-based methods have been acknowledged as a powerful paradigm for end-to-end visuomotor control in robotics. Most existing approaches adopt a Diffusion Policy in U-Net architecture (DP-U), which, while effective, suffers from limited global context modeling and over-smoothing artifacts. To address these issues, we propose U-DiT Policy, a novel U-shaped Diffusion Transformer framework. U-DiT preserves the multi-scale feature fusion advantages of U-Net while integrating the global context modeling capability of Transformers, thereby enhancing representational power and policy expressiveness. We evaluate U-DiT extensively across both simulation and real-world robotic manipulation tasks. In simulation, U-DiT achieves an average performance gain of 10\% over baseline methods and surpasses Transformer-based diffusion policies (DP-T) that use AdaLN blocks by 6\% under comparable parameter budgets. On real-world robotic tasks, U-DiT demonstrates superior generalization and robustness, achieving an average improvement of 22.5\% over DP-U. In addition, robustness and generalization experiments under distractor and lighting variations further highlight the advantages of U-DiT. These results highlight the effectiveness and practical potential of U-DiT Policy as a new foundation for diffusion-based robotic manipulation. |
| title | U-DiT Policy: U-shaped Diffusion Transformers for Robotic Manipulation |
| topic | Robotics |
| url | https://arxiv.org/abs/2509.24579 |