Learning Controlled Separation of Small Objects Between Two Fingers with a Tactile Skin

2026-05-29

AI Summary

This study addresses the challenge of enabling a multi-purpose robotic hand to precisely retain a specified number of tiny objects from a cluttered pile using only tactile sensing, without visual input. The authors propose a reinforcement learning approach leveraging high-resolution tactile skins, integrated with a sparse reward scheme and a contact location estimator. A separation policy trained entirely in simulation is successfully transferred to the real-world DLR-Hand II platform. This work presents the first demonstration of purely tactile-driven, controllable separation of small objects, highlighting the critical role of spatially resolved touch in fine manipulation. In simulation, high-resolution tactile sensing achieves near-perfect performance; even with a coarse 4×4 taxel array, success rates improve by up to 20% compared to joint-sensing-only baselines—a result corroborated in physical experiments.

Abstract

We introduce and solve the novel task of controlled separation of small objects with two fingers of a multi-purpose robotic hand: after grasping into a box of small objects, the task is to drop as many of them until a desired number remains between the fingers. The objects are small compared to the width of the fingers but also in absolute terms. In our case little pellets with a diameter of only 6mm are handled. We show that the task can be performed purely tactile (no vision) using a spatially-resolved tactile skin on a fingertip. The separation policy is trained in simulation via reinforcement learning using a straightforward sparse reward, which basically checks if the desired number of objects is reached. In simulation experiments, we provide an exhaustive analysis of the benefits of using spatially-resolved tactile feedback: while an ideal (high-resolution) tactile sensor allows solving the task almost perfectly, a sensor with lower spatial resolution (here 4x4 taxels) still leads to an improvement of up to 20% compared to using only the fingers' joint sensors. For this analysis, we further train an estimator alongside the policy that predicts the ground truth contact positions. Finally, we demonstrate the successful sim-to-real transfer for the DLR-Hand II equipped with a tactile skin.

Problem

• controlled separation
• small objects
• tactile skin
• robotic hand
• tactile feedback

Innovation

• tactile skin
• controlled separation
• reinforcement learning
• sim-to-real transfer
• spatially-resolved tactile feedback