Tactile Memory with Soft Robot: Robust Object Insertion via Masked Encoding and Soft Wrist

TL;DR

This paper presents TaMeSo-bot, a soft robotic system with tactile memory and a transformer-based model that enables robust, adaptable object insertion by learning rich tactile representations and reusing past demonstrations.

Contribution

It introduces a novel tactile memory system with a masked tactile trajectory transformer that autonomously learns task-relevant features for flexible manipulation.

Findings

MAT$^3$ achieves higher success rates than baselines.

The system adapts effectively to unseen pegs and conditions.

It enables safe contact exploration with a soft wrist.

Abstract

Tactile memory, the ability to store and retrieve touch-based experience, is critical for contact-rich tasks such as key insertion under uncertainty. To replicate this capability, we introduce Tactile Memory with Soft Robot (TaMeSo-bot), a system that integrates a soft wrist with tactile retrieval-based control to enable safe and robust manipulation. The soft wrist allows safe contact exploration during data collection, while tactile memory reuses past demonstrations via retrieval for flexible adaptation to unseen scenarios. The core of this system is the Masked Tactile Trajectory Transformer (MAT$^\text{3}$), which jointly models spatiotemporal interactions between robot actions, distributed tactile feedback, force-torque measurements, and proprioceptive signals. Through masked-token prediction, MAT$^\text{3}$ learns rich spatiotemporal representations by inferring missing sensory information from context, autonomously extracting task-relevant features without explicit subtask segmentation. We validate our approach on peg-in-hole tasks with diverse pegs and conditions in real-robot experiments. Our extensive evaluation demonstrates that MAT$^\text{3}$ achieves higher success rates than the baselines over all conditions and shows remarkable capability to adapt to unseen pegs and conditions.