Sign in Subscribe
Machine learning

Breakthrough in Robotics: Multimodal Sensory Skin Enhances Human-Like Sensing Abilities

Strategic Insights

— 1 min read

Scientists at the University of College London (UCL) have made significant strides in robotics by developing a single-layer multimodal sensory skin that can capture a wide range of data over a large surface area. This innovative skin uses a highly sensitive hydrogel membrane to provide a soft interface, emulating the biological functions of human skin. The research team, led by Thomas George Thuruthel, has successfully demonstrated the skin's ability to predict environmental conditions, localize human touch, and generate proprioceptive data. This breakthrough has the potential to revolutionize the field of robotics and sensitive systems.

Key Takeaways:

  • The human skin-like sensory skin is made using a highly sensitive hydrogel membrane, which is fabricated into a single layer.
  • The skin uses electrical impedance tomography techniques to access up to 863,040 conductive pathways across the membrane, allowing it to identify distinct types of multimodal stimuli.
  • The research team demonstrated the skin's ability to predict environmental conditions, localize human touch, and generate proprioceptive data through comprehensive physical testing.
  • The highly redundant and coupled sensory information from the pathways can be structured using data-driven techniques, selecting which pathways should be monitored for efficient multimodal perception.
  • The skin's versatility is demonstrated by casting it into the shape and size of an adult human hand, showcasing its potential for use in sensitive systems.
  • The research framework addresses the challenge of physically extracting meaningful information in multimodal soft sensing, opening new directions for the information-led design of single-layer skins.

Statistics:

  • The multimodal sensory skin uses up to 863,040 conductive pathways to access multimodal stimuli.
  • The skin is made using a highly sensitive hydrogel membrane, which provides a soft interface.
  • The research team demonstrated the skin's ability to predict environmental conditions with a success rate of unknown (as per the announcement).
  • The skin is capable of localizing human touch and generating proprioceptive data through electrical impedance tomography techniques.
  • The skin's framework is designed to select the most efficient pathways for multimodal perception, utilizing data-driven techniques.

Sources:

  • Multimodal information structuring with single-layer soft skins and high-density electrical impedance tomography. Science Robotics, 2025;10(103).
  • University College London (UCL) Reports Findings in Robotics (Multimodal information structuring with single-layer soft skins and high-density electrical impedance tomography). Robotics & Machine Learning. June 23, 2025; p 3415.

Read more