AI

ViTacPhys Estimates Mass and Stiffness Before It Squeezes

Robb Harlan 4 min read

Most grasp policies see an object and close. ViTacPhys tries to guess mass, friction class, and stiffness first, then conditions the grasp on those numbers.

The paper (arXiv:2608.21355, submitted August 21) and the project page describe a visuo-tactile pipeline trained on 1,800 one-second human demonstrations of 60 rigid and deformable objects.

Lab humanoid with a 6-DoF dexterous hand holding a paper cup while an operator wears a Manus teleoperation glove
In-lab humanoid, 6-DoF hand, Manus glove. Source: ViTacPhys project page.

What it claims to measure

On seen objects, the paper reports:

  • 97.2% mass-class accuracy
  • 98.8% friction-class accuracy
  • stiffness mean absolute percentage error 5.51%

On held-out objects from known categories: 87.5% mass, 97.5% friction, stiffness MAPE 9.08%. One-shot, with very little object diversity, falls off hard (mass accuracy 0.492 in their table).

The grasp policy that uses those predictions, after a human-to-robot transfer step, reaches 95.0% total grasp success on in-distribution objects and 83.4% on out-of-distribution objects. For OOD objects both methods lifted, the force traces sit closer to human teleop than ACT.

Transfer is not magic. They fine-tune with robot teleoperation, visually augmented dexterous-hand demos, and matched-action human demos. The project page says the strongest OOD transfer uses all three.

Grid comparing ACT, ViTacFormer, and ViTacPhys grasps on bottles, cans, and a paper cup
Teaser: ACT and ViTacFormer crush or drop several items; ViTacPhys labels mass/stiffness/friction under each column. Source: arXiv:2608.21355.

How they collected it

Demonstrations are 1 second at 30 Hz, two protocols: vertical lift and lateral shake. Labels come from a precision scale, force–displacement fits, and an inclined-plane friction test. Wrist RGB and fingertip tactile maps are aligned to the robot’s sensors.

The real robot in the paper is a 7-DoF arm with a direct-drive 6-DoF dexterous hand. Fingertips carry piezoresistive arrays matching the human glove. Teleop uses a Quantum Manus glove: 10 demos on each of 50 objects.

Motion-capture cameras, a laptop, fingertip tactile sensors on a human hand, and a matching robot fingertip
Acquisition setup and the human-to-robot sensor match. Source: ViTacPhys project page.

A Human’s Take

This is the right complaint about vision-only grasping. A full bottle and an empty can can look similar until you close. Estimating a stiffness number, even a crude one, is more useful than another VLM caption.

The 95% ID number is a lab table, not a shift. The one-shot collapse is the part I would not skip: if the object class is new, the property head is not a sensor. It is a guess with a nice error bar. Still, I would rather a policy that admits mass and friction exist than one that treats every cup like a brick.

Sources