# Digger Finger: GelSight Tactile Sensor for Object Identification Inside Granular Media

arXiv: https://arxiv.org/abs/2102.10230 · HTML: https://mlsanity.com/paper/2102.10230/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Feb 2021.

## Built on by (most cited first, 8 of 8)

- [Dexterity from Touch: Self-Supervised Pre-Training of Tactile Representations with Robotic Play](https://mlsanity.com/paper/2303.12076/) (2023)
- [See to Touch: Learning Tactile Dexterity through Visual Incentives](https://mlsanity.com/paper/2309.12300/) (2023)
- [DenseTact: Optical Tactile Sensor for Dense Shape Reconstruction](https://mlsanity.com/paper/2201.01367/) (2022)
- [Tactile-based Object Retrieval From Granular Media](https://mlsanity.com/paper/2402.04536/) (2024)
- [Towards Design and Development of an ArUco Markers-Based Quantitative Surface Tactile Sensor](https://mlsanity.com/paper/2310.08398/) (2023)
- [A Modularized Design Approach for GelSight Family of Vision-based Tactile Sensors](https://mlsanity.com/paper/2504.14739/) (2025)
- [TransDex: Pre-training Visuo-Tactile Policy with Point Cloud Reconstruction for Dexterous Manipulation of Transparent Objects](https://mlsanity.com/paper/2603.13869/) (2026)
- [DeCAL: Towards Physically-Grounded Dexterous Vision-Language-Action Models via Contact-Aware Latent Co-Imagination](https://mlsanity.com/paper/2609.09119/) (2026)

## Abstract (arXiv)

In this paper we present an early prototype of the Digger Finger that is designed to easily penetrate granular media and is equipped with the GelSight sensor. Identifying objects buried in granular media using tactile sensors is a challenging task. First, particle jamming in granular media prevents downward movement. Second, the granular media particles tend to get stuck between the sensing surface and the object of interest, distorting the actual shape of the object. To tackle these challenges we present a Digger Finger prototype. It is capable of fluidizing granular media during penetration using mechanical vibrations. It is equipped with high resolution vision based tactile sensing to identify objects buried inside granular media. We describe the experimental procedures we use to evaluate these fluidizing and buried shape recognition capabilities. A robot with such fingers can perform explosive ordnance disposal and Improvised Explosive Device (IED) detection tasks at a much a finer resolution compared to techniques like Ground Penetration Radars (GPRs). Sensors like the Digger Finger will allow robotic manipulation research to move beyond only manipulating rigid objects.