SCANS: A Soft Gripper with Curvature and Spectroscopy Sensors for In-Hand Material Differentiation

TL;DR

SCANS tackles in-hand material differentiation for soft robotics by integrating curvature sensing and spectroscopy in an electronics-free finger. It leverages PMMA-based spectroscopic waveguides to collect in-hand spectra across 800-2500 nm and models curvature as $\kappa = \frac{1}{R}$. Through LDA and SAM analyses across 37 items, the work shows that near-infrared bands carry discriminative information, enabling inter- and intra-class differentiation. The paper offers a complete fabrication workflow, a modular measurement architecture, and publicly available parts list and code, highlighting the potential of optical sensing as a multi-functional modality for safe, adaptable soft manipulation and automated material assessment.

Abstract

We introduce the soft curvature and spectroscopy (SCANS) system: a versatile, electronics-free, fluidically actuated soft manipulator capable of assessing the spectral properties of objects either in hand or through pre-touch caging. This platform offers a wider spectral sensing capability than previous soft robotic counterparts. We perform a material analysis to explore optimal soft substrates for spectral sensing, and evaluate both pre-touch and in-hand performance. Experiments demonstrate explainable, statistical separation across diverse object classes and sizes (metal, wood, plastic, organic, paper, foam), with large spectral angle differences between items. Through linear discriminant analysis, we show that sensitivity in the near-infrared wavelengths is critical to distinguishing visually similar objects. These capabilities advance the potential of optics as a multi-functional sensory modality for soft robots. The complete parts list, assembly guidelines, and processing code for the SCANS gripper are accessible at: https://parses-lab.github.io/scans/.

Figures (8)

• Figure 1: The top three pictures show distinct points during a grasp cycle, where the spectral signal (bottom) is acquired for the ambient environment, caging pre-grasp, and in-hand response.
• Figure 2: (a) Exploded 3D rendering of the components in a single SCANS finger. (b) Operating principle for sensor with waveguide and dual PMMA waveguides for in-hand spectroscopic measurements.
• Figure 3: (a) Optical transmission profiles for each of the candidate materials for the creation of the soft sensors. These readings are converted from raw digital counts by normalizing against the transmission profile of a 6.5 mm piece of uncoated Borosilicate glass. (b) Insertion loss calculated using Eq. \ref{['eq:insertion_loss']} for lengths of PMMA fibers 4-15 mm in length. Fit regression lines for each length show differing loss rates. Lower loss values are desirable, as increased loss corresponds to lower perceived signal.
• Figure 4: Fabrication process for single SCANS waveguide from raw materials to integration in a soft robot finger.
• Figure 5: (a) System architecture used to quantitatively evaluate the SCANS sensor performance. (b) Associated curvature waveguide spectral response in unactuated and actuated states as shown in (c). Note the decrease in spectral intensity. (c) Motion tracking profile of the finger in unactuated and actuated states with the associated curvature measurements in radians.