Scalable, Simulation-Guided Compliant Tactile Finger Design

TL;DR

This work proposes a simulation frame-work for the end-to-end forward design of GelSight Fin Ray sensors and investigates design choices available in the compliant grippers, namely gel pad shapes, illumination conditions, Fin Ray gripper sizes, and Fin Ray stiffness.

Abstract

Compliant grippers enable robots to work with humans in unstructured environments. In general, these grippers can improve with tactile sensing to estimate the state of objects around them to precisely manipulate objects. However, co-designing compliant structures with high-resolution tactile sensing is a challenging task. We propose a simulation framework for the end-to-end forward design of GelSight Fin Ray sensors. Our simulation framework consists of mechanical simulation using the finite element method (FEM) and optical simulation including physically based rendering (PBR). To simulate the fluorescent paint used in these GelSight Fin Rays, we propose an efficient method that can be directly integrated in PBR. Using the simulation framework, we investigate design choices available in the compliant grippers, namely gel pad shapes, illumination conditions, Fin Ray gripper sizes, and Fin Ray stiffness. This infrastructure enables faster design and prototype time frames of new Fin Ray sensors that have various sensing areas, ranging from 48 mm $\times$ \18 mm to 70 mm $\times$ 35 mm. Given the parameters we choose, we can thus optimize different Fin Ray designs and show their utility in grasping day-to-day objects.

Figures (11)

• Figure 1: Simulation-guided Fin Ray finger design: New design of GelSight Fin Ray fingers can be made, based on the results of FEM mechanical simulation and optical simulation. The simulation provides a good prediction about the finger stiffness and tactile sensing performance.
• Figure 2: The family of GelSight Fin Rays: 1a) is our original one, and 1b) is the GelSight Baby Fin Ray. 2) represents a box like Fin Ray, which can be more useful for applying torques to long tool handles, while 3) represents two versions of a longer and medium-width Fin Ray, on which we performed simulations. 3a) represents a stiffer Fin Ray structure, while 3b) represents a softer structure with thinner ribs.
• Figure 3: Fluorescent model and calibration setup: (A) shows a canonical fluorescent material model hua2023efficient. It consists of absorption and reemission spectra whose peaks are separated by Stokes' shift. (B) shows the imaging setup we created to capture the reflectance at the excitation wavelength, $\lambda=450 \text{nm}$ for calibrating fluorescent paints used in GelSight Fin Ray.
• Figure 4: Fluorescent paint fit: This shows the comparison of measured emission spectra and simulated emission spectra using a 4D parametric model for red fluorescent paint (left) and green fluorescent paint (right).
• Figure 5: Fluorescent rendering: This image shows the visual of the efficient rendering of fluorescent paint lights using our parametric reflectance model for simulating GelSight Fin Ray sensor.