Compliance and tactile signals improve robot insertion learning success
PHASE: Compliance-Enabled Tactile Phase Retrieval for Few-Shot Insertion Learning
Robotics
Summary
Robot tasks like inserting pegs into holes are hard to teach with just a few examples. The paper shows that when a robot’s wrist is compliant and senses touch and force, it can better understand different stages of the task. This understanding helps the robot use past experiences more effectively to get better at inserting pegs. The authors demonstrate their method improves success rates by a significant margin, especially when starting positions change.
What this means in practice
- •For industrial robot operators: Enable robots to learn precise assembly tasks like peg insertions from few examples by using tactile phase segmentation to improve success rates.
- •For robotics system integrators: Integrate compliance and tactile phase retrieval into assembly robots to enhance robustness against position variation when inserting components.
Authors
Jeremy Siburian, Cristian C. Beltran-Hernandez, Tatsuya Matsushima, Yusuke Iwasawa, Masashi Hamaya, Mai Nishimura
Abstract
Contact-rich assembly tasks such as peg-in-hole insertion remain difficult to learn from limited demonstrations. While retrieval-augmented imitation learning, which augments target demonstrations with relevant prior data, offers a promising direction, its applicability to contact-rich manipulation remains largely unexplored. Contact-rich insertion unfolds over multiple phases from search to insert, and retrieving phase-specific experience from prior data in principled ways remains an open question. Our key insight is that a compliant wrist enables the robot to sustain contact throughout execution, producing rich tactile and force signals that naturally reveal the phase structure of insertion and inform what should be retrieved. Based on this insight, we present PHASE (PHase-Aware Segmentation and REtrieval), a framework for compliance-enabled tactile phase retrieval that integrates multimodal contact-aware representation learning, variable-length phase segmentation from tactile signals, and phase-consistent retrieval for policy learning. We evaluate PHASE on real-world peg-in-hole insertion across five peg geometries, comparing against retrieval strategies drawn from state-of-the-art methods under a shared policy architecture. PHASE improves the overall success rate by 13 percentage points over the strongest non-phase-aware baseline, and improves performance under unseen initial positions by 30 percentage points. These results demonstrate that aligning retrieval with interaction-defined contact phases substantially improves robustness in few-shot insertion learning.