Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2605.12247 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918497783644160 |
|---|---|
| author | Liu, Yibo Oparnica, Stanko Shewchun-Jakaitis, Simon Fu, Guoyi Wang, Jie Yang, Jun Jagannathan, Anand Lo, Tony Hong-Yau |
| author_facet | Liu, Yibo Oparnica, Stanko Shewchun-Jakaitis, Simon Fu, Guoyi Wang, Jie Yang, Jun Jagannathan, Anand Lo, Tony Hong-Yau |
| contents | Contact-rich assembly is fundamental in robotics but poses significant challenges due to uncertainties in relative poses, such as misalignments and small clearances in peg-in-hole tasks. Existing approaches typically address search and high-precision insertion separately, because these tasks involve distinct action patterns. However, supporting both tasks within a single model, without switching models or weights, is desirable for intelligent assembly systems. In this work, we propose SI-Diff, a framework that learns both search and high-precision insertion through a force-domain diffusion policy. To this end, we introduce a new mode-conditioning mechanism that enables the policy to capture distinct action behaviors under a single framework. Moreover, we develop a new search teacher policy that can generate diverse trajectories. By training on successful and efficient demonstrations provided by the teacher policy, the model learns the mapping from tactile and end-effector velocity observations to effective action behaviors. We conduct thorough experiments to show that SI-Diff extends the tolerance to x-y misalignments from 2 mm to 5 mm compared to the state-of-the-art baseline, TacDiffusion, while also demonstrating strong zero-shot transferability to unseen shapes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_12247 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SI-Diff: A Framework for Learning Search and High-Precision Insertion with a Force-Domain Diffusion Policy Liu, Yibo Oparnica, Stanko Shewchun-Jakaitis, Simon Fu, Guoyi Wang, Jie Yang, Jun Jagannathan, Anand Lo, Tony Hong-Yau Robotics Contact-rich assembly is fundamental in robotics but poses significant challenges due to uncertainties in relative poses, such as misalignments and small clearances in peg-in-hole tasks. Existing approaches typically address search and high-precision insertion separately, because these tasks involve distinct action patterns. However, supporting both tasks within a single model, without switching models or weights, is desirable for intelligent assembly systems. In this work, we propose SI-Diff, a framework that learns both search and high-precision insertion through a force-domain diffusion policy. To this end, we introduce a new mode-conditioning mechanism that enables the policy to capture distinct action behaviors under a single framework. Moreover, we develop a new search teacher policy that can generate diverse trajectories. By training on successful and efficient demonstrations provided by the teacher policy, the model learns the mapping from tactile and end-effector velocity observations to effective action behaviors. We conduct thorough experiments to show that SI-Diff extends the tolerance to x-y misalignments from 2 mm to 5 mm compared to the state-of-the-art baseline, TacDiffusion, while also demonstrating strong zero-shot transferability to unseen shapes. |
| title | SI-Diff: A Framework for Learning Search and High-Precision Insertion with a Force-Domain Diffusion Policy |
| topic | Robotics |
| url | https://arxiv.org/abs/2605.12247 |