SI-Diff: A Framework for Learning Search and High-Precision Insertion with a Force-Domain Diffusion Policy

TL;DR

SI-Diff is a unified framework that learns both search and high-precision insertion in contact-rich assembly tasks using a force-domain diffusion policy, improving tolerance to misalignments and enabling zero-shot transfer.

Contribution

It introduces a mode-conditioning mechanism and a search teacher policy to unify search and insertion tasks within a single model, enhancing robustness and transferability.

Findings

Extends x-y misalignment tolerance from 2mm to 5mm.

Demonstrates strong zero-shot transfer to unseen shapes.

Outperforms the state-of-the-art TacDiffusion baseline.

Abstract

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.