×
Meta is releasing new research to advance embodied AI and robotics
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

Advancing embodied AI: Meta FAIR is making significant strides in robotics research, aiming to develop AI systems that can perceive and interact with their surroundings as effectively as humans.

  • Meta is releasing several new research artifacts that focus on touch perception, robot dexterity, and human-robot interaction, which are crucial for achieving advanced machine intelligence (AMI).
  • The company is partnering with GelSight Inc and Wonik Robotics to develop and commercialize tactile sensing innovations, fostering an open ecosystem for AI research.

Breakthrough in touch perception: Meta Sparsh, a new general-purpose touch representation, is being released to enable AI to perceive what’s inaccessible through vision alone.

  • Sparsh is the first encoder that works across various vision-based tactile sensors and tasks, leveraging self-supervised learning to avoid the need for labels.
  • The technology outperforms task- and sensor-specific models by an average of over 95% on a novel benchmark consisting of six touch-centric tasks.

Advanced tactile sensing: Meta Digit 360, a revolutionary artificial fingertip, delivers human-level tactile sensing capabilities.

  • Equipped with over 18 sensing features, Digit 360 can detect miniature spatial changes and capture forces as small as 1 millinewton.
  • The sensor incorporates a wide-field-of-view optical system with over 8 million taxels and multiple sensing modalities to perceive vibrations, heat, and even odors.
  • GelSight Inc will manufacture and distribute Digit 360, making it available for purchase next year.

Integrating tactile sensing: Meta Digit Plexus provides a standardized platform for integrating various tactile sensors on a single robot hand.

  • The platform combines vision-based and skin-based tactile sensors across fingertips, fingers, and palm, enabling seamless data collection, control, and analysis.
  • Wonik Robotics will use Meta Digit Plexus to create a new advanced dexterous robot hand, the next generation of the Allegro Hand.

Enhancing human-robot collaboration: The PARTNR benchmark introduces a framework for evaluating planning and reasoning in human-robot collaboration scenarios.

  • PARTNR comprises 100,000 natural language tasks across 60 houses and over 5,800 unique objects, built on top of the Habitat 3.0 simulator.
  • The benchmark aims to drive advancements in human-robot interaction and collaborative intelligence, transforming AI models from “agents” to “partners.”

Industry partnerships and community engagement: Meta is collaborating with industry leaders to commercialize and distribute its innovations.

  • GelSight Inc will manufacture and distribute Digit 360, with early access available through a call for proposals.
  • Wonik Robotics will develop and distribute the next generation of the Allegro Hand, integrating Meta’s tactile sensing technologies.

Broader implications: Meta’s advancements in robotics and tactile sensing have the potential to impact various fields significantly.

  • These innovations could revolutionize healthcare, manufacturing, supply chains, and energy sectors by enabling more sophisticated and capable robotic systems.
  • The open-source approach and industry partnerships aim to foster community-driven research and accelerate progress in robotics and AI.
  • As AI and robotics continue to evolve, these technologies may pave the way for more efficient, reliable, and adaptable systems that can better serve human needs and preferences.
Advancing embodied AI through progress in touch perception, dexterity, and human-robot interaction

Recent News

‘Agent orchestration’ is the backbone of business ops in the AI era — here’s why

Agent orchestration leverages AI to actively manage interactions and optimize data flow across enterprise systems, promising more responsive and adaptive business environments.

This startup is using AI to help patients decode their X-rays

AI-powered dental imaging system enhances X-rays to improve patient understanding and treatment decisions.

MIT’s latest breakthrough is tiny, but it has big implications for the semiconductor industry

The novel 3D nanoscale transistor design could overcome silicon's physical limitations, potentially leading to more efficient and powerful electronic devices.