UniT: Data Efficient Tactile Representation with Generalization to Unseen Objects

TL;DR

UniT introduces a data-efficient tactile representation learned via VQGAN that generalizes to unseen objects and tasks, enabling zero-shot transfer and improving performance in perception and manipulation tasks.

Contribution

The paper presents UniT, a novel tactile representation learning method using VQGAN that generalizes to unseen objects and tasks, outperforming existing methods in various benchmarks.

Findings

Outperforms existing tactile and visual representation methods.

Enables zero-shot transfer to multiple downstream tasks.

Effective in real-world robotic manipulation scenarios.

Abstract

UniT is an approach to tactile representation learning, using VQGAN to learn a compact latent space and serve as the tactile representation. It uses tactile images obtained from a single simple object to train the representation with generalizability. This tactile representation can be zero-shot transferred to various downstream tasks, including perception tasks and manipulation policy learning. Our benchmarkings on in-hand 3D pose and 6D pose estimation tasks and a tactile classification task show that UniT outperforms existing visual and tactile representation learning methods. Additionally, UniT's effectiveness in policy learning is demonstrated across three real-world tasks involving diverse manipulated objects and complex robot-object-environment interactions. Through extensive experimentation, UniT is shown to be a simple-to-train, plug-and-play, yet widely effective method for tactile representation learning. For more details, please refer to our open-source repository https://github.com/ZhengtongXu/UniT and the project website https://zhengtongxu.github.io/unit-website/.