GelSlim3.0: High-Resolution Measurement of Shape, Force and Slip in a Compact Tactile-Sensing Finger
TL;DR GelSlim 3.0 is a 20 mm-thick, open-source GelSight-style finger whose LEDs and light-shaping lens are optimized in optical simulation so photometric stereo works on a flat gel, giving depth, slip and dense 3D force from one camera. It weighs 45 g, costs about $25 in parts and has a 675 mm² sensing field, twice that of Digit at a comparable thickness.
What they do
Earlier slim GelSlim fingers dropped photometric stereo, because it needs three light colors hitting the gel evenly from different directions, which is hard to fit in a thin finger. GelSlim 3.0 brings it back in a 20 mm-thick finger for a WSG-32 parallel gripper, aimed at bin picking in clutter. One camera gives a high-resolution image, depth, incipient slip and a dense 3D force field, all computed analytically without learning.
How it works
- Optics. A 160° wide-angle Raspberry Pi camera views a flat gel through a hexagonal acrylic lens lit by three LED color channels. LED pose and lens shaping features are optimized in LightTools linked to SolidWorks, with goals of white mixed color, low intensity variance and an illumination centroid at the lens center.
- Build. Ten parts: 3D-printed shells, made-to-order PCB and lens, off-the-shelf camera and screws, plus a cast silicone gel with a laser-cut marker grid. The fingertip snaps on and off.
- Sensing. Marker positions undo the lens distortion. A lookup table calibrated on a sphere plus a Poisson solver gives depth. Slip is flagged when the marker motion in the contact region departs from a rigid-body fit. Marker motion and depth are fed to an FEM gel model to get the force field.
Results
- The initial lens design gave no usable photometric data; the optimized one gives low-noise depth near the center (ball bearing, screw head, screwdriver handle).
- Size 37x80x20 mm, 45 g, 675 mm² sensing field, 640x480 at 90 FPS, about $25 in parts at 1000 units. Digit is 20x27x18 mm with 304 mm²; GelSlim 2.0 is 50x172x25 mm and 222 g.
- Slip detection and force fields are shown qualitatively on a rotated chess piece.
Limitations
Illumination is still not homogeneous, so reconstructions at the edges are noisy. Interreflections in the compact lens can’t be removed, so only fairly large textures are captured, unlike earlier GelSight sensors that resolve micro-geometry. The paper gives no quantitative accuracy for depth, slip or force.
Videos
Figures
Abstract
This work presents a new version of the tactile-sensing finger GelSlim 3.0, which integrates the ability to sense high-resolution shape, force, and slip in a compact form factor for use with small parallel jaw grippers in cluttered bin-picking scenarios. The novel design incorporates the capability to use real-time analytic methods to measure shape, estimate the contact 3D force distribution, and detect incipient slip. To achieve a compact integration, we optimize the optical path from illumination source to camera and other geometric variables in a optical simulation environment. In particular, we optimize the illumination sources and a light shaping lens around the constraints imposed by the photometric stereo algorithm used for depth reconstruction. The optimized optical configuration is integrated into a finger design composed of robust and easily replaceable snap-to-fit fingetip module that allow for ease of manufacture, assembly, use, and repair. To stimulate future research in tactile-sensing and provide the robotics community access to reliable and easily-reproducible tactile finger with a diversity of sensing modalities, we open-source the design and software at https://github.com/mcubelab/gelslim.
Summaries are written automatically from each paper's text. Figures and videos come from the paper and its project page; all credit goes to the authors.