Researchers have created a revolutionary robotic skin that can sense pain, temperature, and pressure, bringing robotic machinery close to human-like touch. The skin is made up of a gel-like material, which transforms the entire surface of a robotic arm into a network of sensitive, intelligent sensors. Traditionally, robotic skins are made out of a patchwork of sensors. Unlike those sensors, this new material can detect pressure, pain, temperature, and even multiple contact points. It fits on a robotic hand like a glove, enabling the hand to detect and gather information about its surroundings.
The researchers, from the University of Cambridge and University College London (UCL), developed a conductive and flexible skin that is easy to fabricate and can be melted down to form different complex shapes. The robots can interact with the physical world in a more meaningful way, with the help of this technology that senses and processes physical inputs. Unlike previous technology, which required multiple sensors embedded in different small areas to feel various kinds of touch, this new electronic “skin” is entirely a sensor, much like our own skin.
The robotic sensor is not as sensitive as human skin but has 860,000 tiny pathways that detect signals like touch and pressure. Sensations like the touch of a finger, hot or cold surface, damage caused by cutting and stabbing, or multiple contact points have been detected by this sensory system. The researchers “taught” the skin to understand which pathway matters the most, utilizing machine learning and combinations of physical tests. This can help the machinery sense different kinds of contact more efficiently.
Electronic skins convert physical information, like pressure or temperature, into electronic ones. In most cases, different sensors are required to understand sensations. These sensors are embedded in soft, flexible material. However, the information from various sensors can interfere with one another, and they are easily damaged.
According to lead author Dr David Hardman from Cambridge’s Department of Engineering, they want to develop a single sensor system that can simultaneously detect multiple types of senses. This type of system is known as multi-modal sensing. Although it is difficult to separate the cause of each signal, multi-modal sensing materials are more robust and are easier to make.
To design the skin, the researchers melted a soft, stretchy gelatine-based hydrogel that is electrically conductive and cast it into the shape of a human hand. Then, the researchers tested different configurations of electrodes to see which configuration gave the most useful information. Thanks to tiny pathways in the conductive material, over 1.7 million pieces of information were generated over the whole hand, from just 32 electrodes placed at the wrist.
The skin was then blasted with a heat gun, gently pressed with fingers, and cut open with a scalpel. The team then used a machine learning model to teach what different touch sensations meant.
“We’re able to squeeze a lot of information from these materials – they can take thousands of measurements very quickly,” said Hardman, who is a postdoctoral researcher in the lab of co-author Professor Fumiya Iida. “They’re measuring lots of different things at once, over a large surface area.”
According to co-author Dr Thomas George Thuruthel from UCL, it is a long way to go before the material becomes as good as human touch; however, the technology is better than anything else present. It is easier to build and can be calibrated to suit the human touch.
Next, the durability of the skin will be put to the test. This technology can bring back sensation to human prosthetics and humanoid robots, which becomes vital in these cases.
Copyright @smorescience. All rights reserved. Do not copy, cite, publish, or distribute this content without permission.
