Meta teams up with GelSight and Wonik Robotics to transform tactile sensors for AI

GelSight will market the Digit360, that digitises touch signals to analyse the modifications in the surroundings
A representational image of robotic arm. — Canva
A representational image of robotic arm. — Canva 

Meta has collaborated with sensor firm GelSight and Wonik Robotics, a South Korean robotics company to advance the commercialisation of tactile sensors for artificial intelligence (AI) applications. It aims to evolve innovative sensing solutions to improve robots’ capability to adjust to the environment, letting them perform a variety of tasks with more precision. 

These latest innovative devices aren’t made up for consumers. Instead, they are developed especially for scientists to help them to utilise these to improve research into AI that can “learn about the world in richer detail” and “better understand and model the physical world.”

Meta refers to the Digit360 as "a tactile fingertip with human-level multimodal sensing capabilities," and its partner, GelSight will market the Digit360, that digitises touch signals to analyse the modifications in the surroundings.

Read more: Meta takes on Google and Bing with revolutionary AI search

According to Meta, “We developed a touch-perception-specific optical system with a wide field of view … for capturing omnidirectional deformations on the fingertip surface,” Meta explained in a blog post. “Additionally, we equipped the sensor with many sensing modalities, since each touch interaction with the environment has a unique profile produced by the mechanical, geometrical, and chemical properties of a surface to perceive vibrations, sense heat, and even smell odour.”

Digit360 will hit the market by next year, and users will receive its early access via Meta’s launched call for proposals. 

Meta’s partnership with Wonik will be concentrated on the latest generation of Wonik’s Allegro Hand, a robotic hand featuring tactile sensors such as Digit360, control boards that encrypt data from tactile sensors onto a host computer, which will be beginning in 2025.