pith. sign in

arxiv: 2511.07797 · v2 · submitted 2025-11-11 · 💻 cs.RO

Characterizing the Resilience and Sensitivity of Polyurethane Vision-Based Tactile Sensors

Benjamin Davis , Hannah Stuart This is my paper

Pith reviewed 2026-05-18 00:19 UTC · model grok-4.3

classification 💻 cs.RO
keywords vision-based tactile sensorspolyurethanesiliconesensor resilienceforce sensitivityrobotic tactile sensingdurability testing
0
0 comments X

The pith

Polyurethane gels make vision-based tactile sensors more durable than silicone for high-load robot tasks.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper tests whether polyurethane rubber can replace silicone gels in vision-based tactile sensors to improve physical durability. It introduces repeatable protocols that measure resilience under normal loading, shear forces, and abrasion, plus direct learning-free checks on force and spatial sensitivity. Results indicate polyurethane versions hold up better overall and operate across a wider range of forces. The main cost is reduced ability to detect very light contacts. A bottle-cap loosening task confirms the lab findings translate to practical use.

Core claim

Polyurethane rubber yields a more robust vision-based tactile sensor than common silicone gels. It sacrifices some sensitivity at low forces yet greatly expands the effective force range, showing clear utility for rugged high-load applications.

What carries the argument

Repeatable characterization protocols for resilience under normal loading, shear, and abrasion, paired with learning-free assessments of force and spatial sensitivity that isolate gel properties from data or model effects.

If this is right

  • Polyurethane VBTSs can sustain higher contact forces and more cycles of loading before failure.
  • These sensors become viable for industrial or outdoor robotic tasks involving repeated wear.
  • The expanded force range allows reliable detection across both moderate and heavy contacts.
  • Real-world demonstrations such as bottle-cap manipulation can be performed without rapid sensor degradation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Material swaps like this one could be tested in other contact-sensing hardware that faces mechanical stress.
  • Formulation tweaks to polyurethane might narrow the low-force sensitivity gap while retaining durability gains.
  • Robotic platforms that currently avoid tactile sensing due to fragility may now incorporate it in demanding settings.

Load-bearing premise

The proposed repeatable characterization protocols for resilience and learning-free sensitivity assessments accurately measure the gels' physical capabilities without confounding effects from sensor design, imaging setup, or material variations.

What would settle it

Repeated high-load abrasion or shear tests in which polyurethane sensors deteriorate at rates comparable to or faster than silicone sensors would falsify the resilience advantage.

Figures

Figures reproduced from arXiv: 2511.07797 by Benjamin Davis, Hannah Stuart.

Figure 1
Figure 1. Figure 1: Our polyurethane VBTS gel is capable of performing grasps without [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: We perform four resilience tests: (A) cyclic compression on an indenter, (B) cyclic local shear on an indenter, (C) cyclic transverse shear on a [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Our model-free spatial sensitivity evaluation uses the sensor reading when pressed onto a ridged surface. The image is preprocessed via background [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Results for cyclic compression (A), cyclic shear on an indenter (B), cyclic shear on a flat surface (C), and abrasion (D) tests across three gel [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Results for force sensitivity (A) and spatial sensitivity (B) tests. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: For demonstration, we perform repeated loosening and tightening [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗
read the original abstract

Vision-based tactile sensors (VBTSs) are a promising technology for robots, providing them with dense signals that can be translated into a multi-faceted understanding of contact. However, existing VBTS tactile surfaces make use of silicone gels, which provide high sensitivity but easily deteriorate from loading and surface wear. We propose that polyurethane rubber, a typically harder material used for high-load applications like shoe soles, rubber wheels, and industrial gaskets, may provide improved physical gel resilience, potentially at the cost of sensitivity. To compare the resilience and sensitivity of two polyurethane gel formulations against a common silicone baseline, we propose a series of repeatable characterization protocols. Our resilience tests assess sensor durability across normal loading, shear loading, and abrasion. For sensitivity, we introduce learning-free assessments of force and spatial sensitivity to directly measure the physical capabilities of each gel without effects introduced from data and model quality. We also include a bottle cap loosening and tightening demonstration to validate the results of our controlled tests with a real-world example. Our results show that polyurethane yields a more robust sensor. While it sacrifices sensitivity at low forces, the effective force range is largely increased, revealing the utility of polyurethane VBTSs over silicone versions in more rugged, high-load applications.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript experimentally compares polyurethane rubber gels to a standard silicone baseline for vision-based tactile sensors (VBTSs). It proposes repeatable protocols for resilience testing under normal loading, shear loading, and abrasion, plus learning-free metrics for force and spatial sensitivity that aim to isolate physical gel properties. Results indicate polyurethane yields greater robustness and a substantially larger effective force range, at the expense of reduced low-force sensitivity, with validation via a bottle-cap loosening/tightening demonstration.

Significance. If the material-specific advantages can be isolated from fabrication and imaging variables, the work would be significant for extending VBTS utility to high-load, rugged robotic tasks such as industrial manipulation and locomotion, where current silicone gels fail due to wear.

major comments (1)
  1. [Resilience and Sensitivity Characterization Protocols] The central claim that polyurethane increases effective force range and robustness rests on the comparison of resilience and learning-free sensitivity protocols. However, the methods description does not report explicit controls or measurements ensuring that gel thickness, marker density/distribution, curing conditions, and camera/lighting geometry were identical (or statistically matched) between polyurethane and silicone specimens. Systematic differences in any of these variables could produce the observed sensitivity and durability differences as artifacts of sensor construction rather than intrinsic material behavior.
minor comments (2)
  1. [Abstract] The abstract states that polyurethane 'largely increased' the effective force range but does not provide the quantitative values or statistical tests supporting this claim; these should be added for clarity.
  2. [Results] Figure captions and axis labels for the sensitivity and durability plots should explicitly note sample sizes, number of trials, and whether error bars represent standard deviation or standard error.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive and detailed review. The feedback highlights an important aspect of experimental rigor that we will address to strengthen the manuscript's claims. We respond to the major comment below.

read point-by-point responses

  1. Referee: [Resilience and Sensitivity Characterization Protocols] The central claim that polyurethane increases effective force range and robustness rests on the comparison of resilience and learning-free sensitivity protocols. However, the methods description does not report explicit controls or measurements ensuring that gel thickness, marker density/distribution, curing conditions, and camera/lighting geometry were identical (or statistically matched) between polyurethane and silicone specimens. Systematic differences in any of these variables could produce the observed sensitivity and durability differences as artifacts of sensor construction rather than intrinsic material behavior.

    Authors: We agree that explicit documentation of fabrication and imaging controls is necessary to attribute observed differences to material properties rather than construction variables. In preparing the specimens, we used the same 3D-printed molds to target consistent gel thickness, applied markers using an identical stamping procedure and density target for both materials, followed the same curing protocol (time and temperature), and employed the same sensor housing, camera module, and LED lighting geometry for all tests. However, the original Methods section did not include quantitative verification of these parameters. In the revised manuscript we will add a new subsection titled 'Specimen Fabrication Controls' that reports: (i) measured gel thickness (mean and standard deviation across n=5 samples per material, obtained via digital caliper and cross-sectional imaging), (ii) marker density (markers per cm^{2} with representative images), (iii) curing conditions (exact duration, temperature, and ambient humidity), and (iv) confirmation that the optical path and illumination remained unchanged across all resilience and sensitivity trials. These additions will allow readers to evaluate the degree of matching directly and will support the claim that differences arise from the intrinsic properties of the polyurethane versus silicone gels. revision: yes

Circularity Check

0 steps flagged

No circularity in experimental comparison of gel materials

full rationale

The paper is a purely empirical study that defines repeatable physical test protocols for resilience (normal loading, shear, abrasion) and learning-free sensitivity metrics, then reports direct measurements on polyurethane versus silicone specimens. No equations, fitted parameters, or predictions appear in the provided text; claims about increased force range and robustness follow from the experimental outcomes rather than reducing to any prior fit or self-referential definition. The protocols are presented as new measurement procedures whose validity rests on physical controls, not on mathematical closure or self-citation chains. This is the expected non-finding for an experimental characterization paper with no derivation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper is experimental characterization work; it relies on standard domain assumptions about material behavior rather than new derivations or entities. No explicit free parameters or invented entities are introduced in the abstract.

axioms (2)
  • domain assumption Silicone gels provide high sensitivity but easily deteriorate from loading and surface wear
    Stated directly as motivation for seeking alternatives in the abstract.
  • domain assumption Polyurethane rubber may provide improved physical gel resilience at potential cost to sensitivity
    Core hypothesis proposed and tested in the abstract.

pith-pipeline@v0.9.0 · 5513 in / 1337 out tokens · 57757 ms · 2026-05-18T00:19:35.602230+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

27 extracted references · 27 canonical work pages · 1 internal anchor

  1. [1]

    Tactile sensing in dexterous robot hands — Review,

    Z. Kappassov, J.-A. Corrales, and V . Perdereau, “Tactile sensing in dexterous robot hands — Review,”Robotics and Autonomous Systems, vol. 74, pp. 195–220, Dec. 2015

    work page 2015

  2. [2]

    M. R. Cutkosky and W. Provancher,Force and Tactile Sensing, pp. 717–736. Cham: Springer International Publishing, 2016

    work page 2016

  3. [3]

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

    M. Lambeta, P.-W. Chou, S. Tian, B. Yang, B. Maloon, V . R. Most, D. Stroud, R. Santos, A. Byagowi, G. Kammerer, D. Jayaraman, and R. Calandra, “DIGIT: A Novel Design for a Low-Cost Compact High- Resolution Tactile Sensor with Application to In-Hand Manipulation,” IEEE Robotics and Automation Letters, vol. 5, pp. 3838–3845, July

    work page

  4. [4]

    arXiv:2005.14679 [cs]

    work page arXiv 2005

  5. [5]

    GelSight: High-Resolution Robot Tactile Sensors for Estimating Geometry and Force,

    W. Yuan, S. Dong, and E. H. Adelson, “GelSight: High-Resolution Robot Tactile Sensors for Estimating Geometry and Force,”Sensors, vol. 17, p. 2762, Dec. 2017. Publisher: Multidisciplinary Digital Publishing Institute

    work page 2017

  6. [6]

    Hardware Technology of Vision-Based Tactile Sensor: A Review,

    S. Zhang, Z. Chen, Y . Gao, W. Wan, J. Shan, H. Xue, F. Sun, Y . Yang, and B. Fang, “Hardware Technology of Vision-Based Tactile Sensor: A Review,”IEEE Sensors Journal, vol. 22, pp. 21410–21427, Nov

    work page

  7. [7]

    Conference Name: IEEE Sensors Journal

    work page

  8. [8]

    DelTact: A Vision-Based Tactile Sensor Using a Dense Color Pattern,

    G. Zhang, Y . Du, H. Yu, and M. Y . Wang, “DelTact: A Vision-Based Tactile Sensor Using a Dense Color Pattern,”IEEE Robotics and Automation Letters, vol. 7, pp. 10778–10785, Oct. 2022

    work page 2022

  9. [9]

    A Novel Dynamic-Vision-Based Approach for Tactile Sensing Applications,

    F. Baghaei Naeini, A. M. AlAli, R. Al-Husari, A. Rigi, M. K. Al- Sharman, D. Makris, and Y . Zweiri, “A Novel Dynamic-Vision-Based Approach for Tactile Sensing Applications,”IEEE Transactions on Instrumentation and Measurement, vol. 69, pp. 1881–1893, May 2020

    work page 2020

  10. [10]

    Minsight: A Fingertip-Sized Vision-Based Tactile Sensor for Robotic Manipula- tion,

    I. Andrussow, H. Sun, K. J. Kuchenbecker, and G. Martius, “Minsight: A Fingertip-Sized Vision-Based Tactile Sensor for Robotic Manipula- tion,”Advanced Intelligent Systems, vol. 5, no. 8, p. 2300042, 2023

    work page 2023

  11. [11]

    DTact: A Vision-Based Tactile Sensor that Measures High-Resolution 3D Geometry Directly from Darkness,

    C. Lin, Z. Lin, S. Wang, and H. Xu, “DTact: A Vision-Based Tactile Sensor that Measures High-Resolution 3D Geometry Directly from Darkness,” Sept. 2022. arXiv:2209.13916 [cs]

    work page arXiv 2022

  12. [12]

    Vision-Based Tactile Sensor Mechanism for the Estimation of Contact Position and Force Distribution Using Deep Learning,

    V . Kakani, X. Cui, M. Ma, and H. Kim, “Vision-Based Tactile Sensor Mechanism for the Estimation of Contact Position and Force Distribution Using Deep Learning,”Sensors, vol. 21, p. 1920, Jan

    work page 1920

  13. [13]

    Publisher: Multidisciplinary Digital Publishing Institute

    work page

  14. [14]

    FingerVision Tactile Sensor Design and Slip Detection Using Convolutional LSTM Network

    Y . Zhang, Z. Kan, Y . A. Tse, Y . Yang, and M. Y . Wang, “FingerVision Tactile Sensor Design and Slip Detection Using Convolutional LSTM Network,” Oct. 2018. arXiv:1810.02653 [cs]

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  15. [15]

    PolyTouch: A Robust Multi-Modal Tactile Sensor for Contact- rich Manipulation Using Tactile-Diffusion Policies,

    J. Zhao, N. Kuppuswamy, S. Feng, B. Burchfiel, and E. Adelson, “PolyTouch: A Robust Multi-Modal Tactile Sensor for Contact- rich Manipulation Using Tactile-Diffusion Policies,” Apr. 2025. arXiv:2504.19341 [cs]

    work page arXiv 2025

  16. [16]

    GelSight Wedge: Measuring High-Resolution 3D Contact Geometry with a Compact Robot Finger,

    S. Wang, Y . She, B. Romero, and E. Adelson, “GelSight Wedge: Measuring High-Resolution 3D Contact Geometry with a Compact Robot Finger,” in2021 IEEE International Conference on Robotics and Automation (ICRA), pp. 6468–6475, May 2021. ISSN: 2577- 087X

    work page 2021

  17. [17]

    High-Speed Tactile Braille Reading via Biomimetic Sliding Interactions,

    P. Potdar, D. Hardman, E. Almanzor, and F. Iida, “High-Speed Tactile Braille Reading via Biomimetic Sliding Interactions,”IEEE Robotics and Automation Letters, vol. 9, pp. 2614–2621, Mar. 2024

    work page 2024

  18. [18]

    Design and development of a robust vision-based tactile sensor,

    P. Rayamane, Z. Ji, and M. Packianather, “Design and development of a robust vision-based tactile sensor,” in2022 IEEE/ASME Inter- national Conference on Advanced Intelligent Mechatronics (AIM), pp. 1417–1423, July 2022. ISSN: 2159-6255

    work page 2022

  19. [19]

    Combining finger vision and optical tactile sensing: Reducing and handling errors while cutting vegetables,

    A. Yamaguchi and C. G. Atkeson, “Combining finger vision and optical tactile sensing: Reducing and handling errors while cutting vegetables,” in2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids), (Cancun, Mexico), pp. 1045–1051, IEEE, Nov. 2016

    work page 2016

  20. [20]

    The TIP Benchmark: A Tactile Image- Based Psychophysics-Inspired Benchmark for Artificial Tactile Sen- sors,

    T. Liu and B. Ward-Cherrier, “The TIP Benchmark: A Tactile Image- Based Psychophysics-Inspired Benchmark for Artificial Tactile Sen- sors,” inHaptics: Understanding Touch; Technology and Systems; Applications and Interaction(H. Kajimoto, P. Lopes, C. Pacchierotti, C. Basdogan, M. Gori, B. Lemaire-Semail, and M. Marchal, eds.), (Cham), pp. 94–106, Springer...

    work page 2025

  21. [21]

    Sparsh: Self-supervised touch representations for vision- based tactile sensing.arXiv preprint arXiv:2410.24090, 2024

    C. Higuera, A. Sharma, C. K. Bodduluri, T. Fan, P. Lancaster, M. Kalakrishnan, M. Kaess, B. Boots, M. Lambeta, T. Wu, and M. Mukadam, “Sparsh: Self-supervised touch representations for vision-based tactile sensing,” Oct. 2024. arXiv:2410.24090 [cs]

    work page arXiv 2024

  22. [22]

    Tactile mnist: Benchmarking active tactile perception.arXiv preprint arXiv:2506.06361,

    T. Schneider, G. Duret, C. d. Farias, R. Calandra, L. Chen, and J. Peters, “Tactile MNIST: Benchmarking Active Tactile Perception,” June 2025. arXiv:2506.06361 [cs]

    work page arXiv 2025

  23. [23]

    Tactile spatial resolution. I. Two- point discrimination, gap detection, grating resolution, and letter recognition,

    K. O. Johnson and J. R. Phillips, “Tactile spatial resolution. I. Two- point discrimination, gap detection, grating resolution, and letter recognition,”Journal of Neurophysiology, vol. 46, pp. 1177–1192, Dec. 1981. Publisher: American Physiological Society

    work page 1981

  24. [24]

    The limit of tactile spatial resolution in humans,

    R. W. Van Boven and K. O. Johnson, “The limit of tactile spatial resolution in humans,”Neurology, vol. 44, pp. 2361–2361, Dec. 1994. Publisher: Wolters Kluwer

    work page 1994

  25. [25]

    Grating orientation as a measure of tactile spatial acuity,

    J. C. Craig, “Grating orientation as a measure of tactile spatial acuity,”Somatosensory & Motor Research, vol. 16, pp. 197–206, Jan. 1999. Publisher: Taylor & Francis eprint: https://doi.org/10.1080/08990229970456

    work page doi:10.1080/08990229970456 1999

  26. [26]

    Evidence to support the mechanical advantage hypothesis of grasping at low force levels,

    B. Rajakumar and S. K. M. Varadhan, “Evidence to support the mechanical advantage hypothesis of grasping at low force levels,” Scientific Reports, vol. 12, p. 20834, Dec. 2022. Publisher: Nature Publishing Group

    work page 2022

  27. [27]

    A statistical review of industrial robotic grippers,

    L. Birglen and T. Schlicht, “A statistical review of industrial robotic grippers,”Robotics and Computer-Integrated Manufacturing, vol. 49, pp. 88–97, Feb. 2018

    work page 2018