In-Hand Singulation and Scooping Manipulation with a 5 DOF Tactile Gripper

TL;DR

The paper addresses dexterous in-hand manipulation with a compact two-finger gripper that has five degrees of freedom ($5$-DoF) and integrates a GelSight tactile sensor. It combines a tactile-MPC controller for stable grasping with a dedicated tactile-based scooping and insertion strategy to perform singulation of objects buried in granular media and precise credit-card insertion in confined spaces. Key contributions include a novel gripper design, a model-based tactile control for singulation, and a tactile-driven scooping/insertion pipeline, achieving a 76% rub-and-strip success overall, 94.3% for sphere-like objects, and 100% card insertion in tests. The work demonstrates that rich tactile feedback can enable robust in-hand manipulation with limited mechanical DoF, offering practical capabilities for dense, restricted environments.

Abstract

Manipulation tasks often require a high degree of dexterity, typically necessitating grippers with multiple degrees of freedom (DoF). While a robotic hand equipped with multiple fingers can execute precise and intricate manipulation tasks, the inherent redundancy stemming from its extensive DoF often adds unnecessary complexity. In this paper, we introduce the design of a tactile sensor-equipped gripper with two fingers and five DoF. We present a novel design integrating a GelSight tactile sensor, enhancing sensing capabilities and enabling finer control during specific manipulation tasks. To evaluate the gripper's performance, we conduct experiments involving two challenging tasks: 1) retrieving, singularizing, and classification of various objects embedded in granular media, and 2) executing scooping manipulations of credit cards in confined environments to achieve precise insertion. Our results demonstrate the efficiency of the proposed approach, with a high success rate for singulation and classification tasks, particularly for spherical objects at high as 94.3%, and a 100% success rate for scooping and inserting credit cards.

Figures (6)

• Figure 1: In-hand singulation manipulation using the proposed 5 DoF tactile gripper: rubbing off the residual adhesions from the granular media between the finger and the grasped artificial strawberry
• Figure 2: The proposed tactile gripper with five degrees of freedom, composing of a linear actuator and a rotation servo on each finger. One GelSight Mini tactile sensor is mounted one the left fingertip.
• Figure 3: Model for the designed scooping motion for the proposed gripper using a 3D printed fingernail
• Figure 4: Sequential images of in-hand grasping and singulation manipulation of a peanut buried in rice. The green arrow in (b) indicates the motion of the gripper, while the yellow arrow represents the motion of the robot arm. The same arrow notation is used in Figure \ref{['fig:cardexp']}(b).
• Figure 5: Selected objects for singulation task and the classification result.