DIGIT: A Novel Design for a Low-Cost Compact High-Resolution Tactile Sensor With Application to In-Hand Manipulation

Abstract

Despite decades of research, general purpose inhand manipulation remains one of the unsolved challenges of robotics. One of the contributing factors that limit current robotic manipulation systems is the difficulty of precisely sensing contact forces -sensing and reasoning about contact forces are crucial to accurately control interactions with the environment. As a step towards enabling better robotic manipulation, we introduce DIGIT, an inexpensive, compact, and high-resolution tactile sensor geared towards in-hand manipulation. DIGIT improves upon past vision-based tactile sensors by miniaturizing the form factor to be mountable on multi-fingered hands, and by providing several design improvements that result in an easier, more repeatable manufacturing process, and enhanced reliability. We demonstrate the capabilities of the DIGIT sensor by training deep neural network model-based controllers to manipulate glass marbles in-hand with a multi-finger robotic hand. To provide the robotic community access to reliable and low-cost tactile sensors, we open-source the DIGIT design at www.digit.ml.

A Differentiable Extended Kalman Filter for Object Tracking Under Sliding Regime

“…More recent works exploit the availability of vision-based tactile sensors ( Yuan et al, 2017 ; Lambeta et al, 2020 ). These sensors provide rich contact information in the form of RGB images capturing the local deformation of soft materials covering the sensor itself.…”

“…Given the large availability of neural and deep neural modules for images processing, several works have combined learning techniques with the high dimensional data provided by these kind of sensors. Sodhi et al (2020) use the Digit ( Lambeta et al, 2020 ) vision-based tactile sensor to estimate the pose of a planar object being pushed by a manipulator. They first learn a suitable observation model that maps consecutive tactile images to the relative pose of the sensor.…”

Blocks World of Touch: Exploiting the Advantages of All-Around Finger Sensing in Robot Grasping

“…Markers were also added to the membrane of the GelSight sensor, enabling applying the same set of methods that were explored in the TacTip sensors. There are some other sensor designs and adaptations for robotic fingers in Yuan et al ( 2017 ), Donlon et al ( 2018 ), and Lambeta et al ( 2020 ). In Yuan et al ( 2017 ), matte aluminum powder is used for improved surface reconstruction, together with the LEDs being placed next to the elastomer, and the elastomer being slightly curved on the top/external side.…”

“…A stretchy textured fabric is also placed on top of the tactile membrane to prevent damages to the elastomer and to improve tactile signal strength. In Lambeta et al ( 2020 ), the authors propose a compact design, with a USB “plug-and-play” port and an easily replaceable elastomer, secured with a single screw mount.…”

Robust Estimation of Contact Force and Location for Magnetic-Field-Based Soft Tactile Sensor Considering Magnetic Source Inconsistency

“…At present, flexible tactile sensors draw from the following measuring principle, including capacitive [ 7 ], resistive [ 8 ], piezoelectric [ 9 ], optical [ 10 ], magnetic [ 11 ], vision-based [ 12 , 13 ], etc. The tactile sensors based on capacitive, resistive, and piezoelectric were generally combined in array to expand the sensing area; the pitch of the taxel (tactile pixel) limits the spatial resolution of such tactile sensors.…”