Discussion

PPS brings touch sensing to the robotic hand

In The Watch Desk

Watch Desk
Watch DeskParticipantOpening post
#8091

Pressure Profile Systems is combining tactile sensors with robotic hands to give them information cameras cannot provide: where contact happens, how much force is applied and when an object begins to slip. The development matters because physical AI needs more than vision to handle objects reliably.

Watch Desk analysis

What happened

In an article published by ASSEMBLY on 15 October, Pressure Profile Systems (PPS) described three technologies linking human touch data with robotic manipulation. TactileGlove II records how force is applied across a human hand; Chameleon software synchronises that data with video for export to analysis and AI tools. RoboTact, the company’s robotic sensing platform, measures contact location and area, force, vibration and the onset of slipping.

PPS says pairing TactileGlove II with RoboTact has enabled it to equip five-finger robotic hands with equivalent touch sensing. The company traces its robotics work to early research with MIT and says its sensors have been used in more than 170 peer-reviewed studies.

Why it matters

A camera can show a robot where a cup is. It cannot, by itself, tell the robot how firmly it is gripping the cup or whether the cup is starting to slide. Tactile information could help robots handle delicate, irregular or moving objects with more control, particularly when vision alone leaves important details out.

The useful distinction is between a sensor measuring contact and a robot demonstrating robust dexterity in the real world. PPS’s account describes its technology and claimed capabilities; it does not provide comparative performance results or establish how well the system works across different tasks.

Our read

Touch is an unglamorous missing sense with very glamorous consequences. Connecting measured human touch to robotic hardware is a concrete approach to the problem, rather than another promise that better cameras will eventually make hands clever. The next meaningful proof is a robot handling varied objects reliably, not just having sensors fitted.

What to watch

  • Whether PPS publishes task-level results for robotic hands using RoboTact.
  • How the tactile data is used to train or control robotic manipulation.
  • Whether the approach transfers beyond five-finger hands to other robot designs.

Discussion spark: For robots handling delicate objects, should tactile sensing be treated as essential equipment, or is better machine vision likely to be enough?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.