The ideal approach in robotics and prosthetics is based on imitating the sensorimotor abilities of humans, for example those of a hand. In robotics, force and torque sensors are usually permanently integrated. These measuring cells provide valuable feedback on the interaction of the robotic system - such as an artificial hand - with the environment. However, traditional sensors reach their limits when it comes to individual adaptations. In addition, they cannot be attached to arbitrary objects. In other words: until now, there has been no process that could be used to produce sensors for rigid objects of any shape and size.

The research team in the MIRMI laboratory (from left to right): Dr. Amartya Ganguly, Sonja Groß, Diego Hidalgo-Carvajal

© TUM

This is where the research of Sonja Groß and Diego Hidalgo, scientists from the Munich Institute of Robotics and Machine Intelligence (MIRMI) at the Technical University of Munich (TUM), comes in, which they have just presented at the ICRA robotics conference in London. The special feature: A soft skin-like material that nestles around the objects. The research group has also developed a 'framework' to produce this skin largely automatically: "We use software to build the structure for the sensors," says Hidalgo, "and we send this information to a 3D printer, where our soft sensors are produced." The printer applies a conductive, black paste to liquid silicone. While the silicone hardens, the paste remains liquid and is enclosed by the silicone. If the sensors are pressed or stretched, their electrical resistance changes. "This tells us how hard a surface has been pressed or stretched. We use this principle to understand interactions with objects in general and, more specifically, to control an artificial hand that interacts with these objects," explains Hidalgo. The sensors embedded in silicone adapt to the respective surface (such as fingers or hands) and still provide reliably precise data that can be used for interaction with the environment.

"The integration of these soft, skin-like sensors into 3D objects opens up new avenues for advanced haptics in artificial intelligence," says Executive Director of MIRMI Prof. Sami Haddadin with conviction. This is because the sensors provide valuable information about compressive forces and deformations in real time, for example, giving immediate feedback. In this way, they expand the perception of an object or a robotic hand and enable more sophisticated and sensitive interaction. Haddadin ventures a prediction: "This work has the potential to revolutionize industries such as robotics, prosthetics and human-machine interaction in general by enabling wireless and customizable sensorization for any object and machine."