Falcon: Fast Visuomotor Policies via Partial Denoising
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912404912209920 |
|---|---|
| author | Chen, Haojun Liu, Minghao Ma, Chengdong Ma, Xiaojian Ma, Zailin Wu, Huimin Chen, Yuanpei Zhong, Yifan Wang, Mingzhi Li, Qing Yang, Yaodong |
| author_facet | Chen, Haojun Liu, Minghao Ma, Chengdong Ma, Xiaojian Ma, Zailin Wu, Huimin Chen, Yuanpei Zhong, Yifan Wang, Mingzhi Li, Qing Yang, Yaodong |
| contents | Diffusion policies are widely adopted in complex visuomotor tasks for their ability to capture multimodal action distributions. However, the multiple sampling steps required for action generation significantly harm real-time inference efficiency, which limits their applicability in real-time decision-making scenarios. Existing acceleration techniques either require retraining or degrade performance under low sampling steps. Here we propose Falcon, which mitigates this speed-performance trade-off and achieves further acceleration. The core insight is that visuomotor tasks exhibit sequential dependencies between actions. Falcon leverages this by reusing partially denoised actions from historical information rather than sampling from Gaussian noise at each step. By integrating current observations, Falcon reduces sampling steps while preserving performance. Importantly, Falcon is a training-free algorithm that can be applied as a plug-in to further improve decision efficiency on top of existing acceleration techniques. We validated Falcon in 48 simulated environments and 2 real-world robot experiments. demonstrating a 2-7x speedup with negligible performance degradation, offering a promising direction for efficient visuomotor policy design. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_00339 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Falcon: Fast Visuomotor Policies via Partial Denoising Chen, Haojun Liu, Minghao Ma, Chengdong Ma, Xiaojian Ma, Zailin Wu, Huimin Chen, Yuanpei Zhong, Yifan Wang, Mingzhi Li, Qing Yang, Yaodong Robotics Diffusion policies are widely adopted in complex visuomotor tasks for their ability to capture multimodal action distributions. However, the multiple sampling steps required for action generation significantly harm real-time inference efficiency, which limits their applicability in real-time decision-making scenarios. Existing acceleration techniques either require retraining or degrade performance under low sampling steps. Here we propose Falcon, which mitigates this speed-performance trade-off and achieves further acceleration. The core insight is that visuomotor tasks exhibit sequential dependencies between actions. Falcon leverages this by reusing partially denoised actions from historical information rather than sampling from Gaussian noise at each step. By integrating current observations, Falcon reduces sampling steps while preserving performance. Importantly, Falcon is a training-free algorithm that can be applied as a plug-in to further improve decision efficiency on top of existing acceleration techniques. We validated Falcon in 48 simulated environments and 2 real-world robot experiments. demonstrating a 2-7x speedup with negligible performance degradation, offering a promising direction for efficient visuomotor policy design. |
| title | Falcon: Fast Visuomotor Policies via Partial Denoising |
| topic | Robotics |
| url | https://arxiv.org/abs/2503.00339 |