Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.06266 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912616662695936 |
|---|---|
| author | Li, Zezeng Yang, Rui Chen, Ruochen Luo, ZhongXuan Chen, Liming |
| author_facet | Li, Zezeng Yang, Rui Chen, Ruochen Luo, ZhongXuan Chen, Liming |
| contents | Diffusion policies have recently emerged as a powerful class of visuomotor controllers for robot manipulation, offering stable training and expressive multi-modal action modeling. However, existing approaches typically treat action generation as an unconstrained denoising process, ignoring valuable a priori knowledge about geometry and control structure. In this work, we propose the Adaptive Diffusion Policy (ADP), a test-time adaptation method that introduces two key inductive biases into the diffusion. First, we embed a geometric manifold constraint that aligns denoising updates with task-relevant subspaces, leveraging the fact that the relative pose between the end-effector and target scene provides a natural gradient direction, and guiding denoising along the geodesic path of the manipulation manifold. Then, to reduce unnecessary exploration and accelerate convergence, we propose an analytically guided initialization: rather than sampling from an uninformative prior, we compute a rough registration between the gripper and target scenes to propose a structured initial noisy action. ADP is compatible with pre-trained diffusion policies and requires no retraining, enabling test-time adaptation that tailors the policy to specific tasks, thereby enhancing generalization across novel tasks and environments. Experiments on RLBench, CALVIN, and real-world datasets show that ADPro, an implementation of ADP, improves success rates, generalization, and sampling efficiency, achieving up to 25% faster execution and 9% points over strong diffusion baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_06266 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ADPro: a Test-time Adaptive Diffusion Policy via Manifold-constrained Denoising and Task-aware Initialization for Robotic Manipulation Li, Zezeng Yang, Rui Chen, Ruochen Luo, ZhongXuan Chen, Liming Robotics Diffusion policies have recently emerged as a powerful class of visuomotor controllers for robot manipulation, offering stable training and expressive multi-modal action modeling. However, existing approaches typically treat action generation as an unconstrained denoising process, ignoring valuable a priori knowledge about geometry and control structure. In this work, we propose the Adaptive Diffusion Policy (ADP), a test-time adaptation method that introduces two key inductive biases into the diffusion. First, we embed a geometric manifold constraint that aligns denoising updates with task-relevant subspaces, leveraging the fact that the relative pose between the end-effector and target scene provides a natural gradient direction, and guiding denoising along the geodesic path of the manipulation manifold. Then, to reduce unnecessary exploration and accelerate convergence, we propose an analytically guided initialization: rather than sampling from an uninformative prior, we compute a rough registration between the gripper and target scenes to propose a structured initial noisy action. ADP is compatible with pre-trained diffusion policies and requires no retraining, enabling test-time adaptation that tailors the policy to specific tasks, thereby enhancing generalization across novel tasks and environments. Experiments on RLBench, CALVIN, and real-world datasets show that ADPro, an implementation of ADP, improves success rates, generalization, and sampling efficiency, achieving up to 25% faster execution and 9% points over strong diffusion baselines. |
| title | ADPro: a Test-time Adaptive Diffusion Policy via Manifold-constrained Denoising and Task-aware Initialization for Robotic Manipulation |
| topic | Robotics |
| url | https://arxiv.org/abs/2508.06266 |