Skip to content
arXiv Robotics — research abstracts· Metin Alp Dogan, Edward Sun, Feng Xu, Daniel Wu, Allen Peng, Dennis Hong, Yuchen Cui·· 1 days agoEditorial score65

Visible Touch: Rendering Contact for Visuomotor Policies

Visible Touch: Rendering Contact for Visuomotor Policies

Summary

This research introduces Visible Touch, a method that integrates contact information into visuoremotor policies by exposing contact signals in the same spatial frame as the scene. The approach uses a custom low-cost magnetic contact sensor and demonstrates improvements in task success across benchmark environments, including LIBERO and real-world contact-rich tasks.

Source: arXiv Robotics — research abstracts · Read original article ↗

Loading article text…

What the source reports

Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.

Reported numbers

  • percentage points

    15.7

    View original evidence
    Visible Touch improves BC-Transformer success by 15.7 percentage points on average in the 2-view setting
    Open source S4
  • percentage points

    25

    View original evidence
    miniVLA on LIBERO gains 25 percentage points on average
    Open source S5
  • percentage points

    30

    View original evidence
    $\pi_{0.5}$ on four real-world contact-rich tasks gains 30 percentage points with our custom sensor
    Open source S5
Source excerpts and review record

Automatically extracted; no manual editorial approval recorded.

y without architectural changes. We operationalize this insight in Visible Touch, paired with a custom low-cost magnetic contact sensor that is open-sourced and fabricated from off-the-shelf parts via a parametric CAD-to-mold pipeline. Across the LIBERO benchmark, Visible Touch improves BC-Transformer success by 15.7 percentage points on average in the 2-view setting, with similar gains in the 1-view setting; control

Open source S4

led comparisons show that the contact-integration strategy strongly affects how effectively tactile information is used. The pattern holds when fine-tuning pretrained VLAs: miniVLA on LIBERO gains 25 percentage points on average, and $\pi_{0.5}$ on four real-world contact-rich tasks gains 30 percentage points with our custom sensor.

Open source S5

Source:arXiv Robotics — research abstracts · arxiv.org

Timezone · UTC

Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).