ROAILGMay 12, 2024

AnyRotate: Gravity-Invariant In-Hand Object Rotation with Sim-to-Real Touch

arXiv:2405.07391v353 citationsh-index: 36CoRL
Originality Incremental advance
AI Analysis

This work addresses the problem of robotic dexterity in manipulation for robotics, representing an incremental advance by applying sim-to-real methods to tactile sensing.

The paper tackles the challenge of achieving gravity-invariant multi-axis in-hand object rotation for robot hands by presenting AnyRotate, a system that uses dense featured sim-to-real touch to train a unified policy, resulting in zero-shot transfer to unseen objects with arbitrary rotation axes and improved robustness through reactive behavior.

Human hands are capable of in-hand manipulation in the presence of different hand motions. For a robot hand, harnessing rich tactile information to achieve this level of dexterity still remains a significant challenge. In this paper, we present AnyRotate, a system for gravity-invariant multi-axis in-hand object rotation using dense featured sim-to-real touch. We tackle this problem by training a dense tactile policy in simulation and present a sim-to-real method for rich tactile sensing to achieve zero-shot policy transfer. Our formulation allows the training of a unified policy to rotate unseen objects about arbitrary rotation axes in any hand direction. In our experiments, we highlight the benefit of capturing detailed contact information when handling objects of varying properties. Interestingly, we found rich multi-fingered tactile sensing can detect unstable grasps and provide a reactive behavior that improves the robustness of the policy. The project website can be found at https://maxyang27896.github.io/anyrotate/.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes