Scientists Develop Robotic Skin That Enables Robots to Feel Touch and Temperature

Scientists have developed a new type of robotic skin that could give robots the ability to feel physical contact. The material can sense pressure, temperature changes, and even damage, such as a sharp cut. These new developments open the door for more responsive and sensitive machines.

The soft, jelly-like skin is made from a gel that can conduct electricity. It can be molded into different shapes, making it suitable for various surfaces. When paired with special electrodes, the skin picks up signals from over 860,000 tiny pathways.

Each pathway responds to different kinds of touch, such as poking, pressing, or heating.

Applications in robotics, medicine, and rescue work

According to the researchers, this breakthrough could support future developments in humanoid robots and artificial limbs, where the sense of touch is essential. The material may also be useful in cars or in emergencies where machines need to handle delicate objects. The study was published in Science Robotics.

Touch sensing is now considered a major step forward in robotics. Scientists are working to build machines that can interact with the world in ways similar to how humans respond through touch.

One sensor replaces many

Traditional electronic skins typically utilize separate sensors for each type of sensation: one for heat, another for pressure, etc. However, these multiple sensors often interfere with one another. Additionally, the materials used to hold them, such as silicone or rubber, are prone to tearing.

Gelatin-Based E-Skin Mimics Human Touch with Single-Material Sensing
by @TheHumanoidHub#Robotics #EmergingTech #Technology #Innovation #TechForGood

cc: @kirkdborne @gp_pulipaka @yuhelenyu pic.twitter.com/W4TZ8ZeUZk

— Ronald van Loon (@Ronald_vanLoon) July 1, 2025

The new material offers a simpler solution. It uses a single type of sensor that can detect multiple types of input. While it remains difficult to separate the signals and identify their exact cause, researchers say the design is easier to build, stronger than previous versions, and cheaper to produce. These benefits could make the skin suitable for large-scale use.

Testing reveals high data accuracy

To test the material, scientists melted the gel and shaped it into a model of a human hand. They attached various electrode setups and ran a series of tests to determine which setup provided the most accurate touch data.

The testing process included exposing the hand to high heat, pressing it with fingers and robotic arms, and even cutting it with a scalpel. Through these experiments, the researchers collected more than 1.7 million data points from the skin’s conductive network.

Machine learning improves future potential

This data was used to train a machine learning system, which could eventually allow robots to recognize and respond to different types of touch.

“We’re not quite at the level where the robotic skin is as good as human skin, but we think it’s better than anything else out there at the moment,” said Thomas George Thuruthel, a lecturer in robotics and artificial intelligence at University College London.

He added that the approach is more adaptable and easier to build than older sensor systems, and can be fine-tuned using real human touch.

Πηγή