An automatic process for creating soft sensors has been developed by scientists at the Munich Institute of Robotics and Machine Intelligence (MIRMI) at the Technical University of Munich (TUM).
The sensor skin is very flexible and can be attached to many surfaces, including fingers, for example. Image Credit: Andreas Heddergott / Technical University of Munich
Such universal measurement cells could be fixed to nearly any kind of object. Applications are expected in robotics and prosthetics.
Detecting and sensing our environment is essential for understanding how to interact with it effectively. This determines how we can perform certain tasks.
Sonja Groß, Researcher, Munich Institute of Robotics and Machine Intelligence, Technical University of Munich
A significant factor in interactions with objects is their shape.
The physical properties of objects, like their flexibility and hardness, impact how one can grasp and manipulate them.
Artificial Hand: Interaction with the Robotic System
The holy grail in prosthetics and robotics is a realistic emulation of the sensorimotor skills of a person like those present in a human hand. As far as robotics is concerned, force and torque sensors have been completely integrated into the majority of the devices.
Valuable feedback has been offered by these measurement sensors on the interactions of the robotic system, like an artificial hand, with its surroundings.
But traditional sensors have been restricted in relation to customization possibilities. Nor could they be fixed to arbitrary objects. In short: so far, no process lasted for producing sensors for firm objects of arbitrary sizes and shapes.
New Framework for Soft Sensors Presented for the First Time
This was considered to be the starting point for the research of Sonja Groß and Diego Hidalgo, which they have currently presented at the ICRA robotics conference in London. The difference: a soft, skin-like material that wraps around objects.
Also, the research group has come up with a framework that mostly automates the production process for this skin.
Hidalgo says, “We use software to build the structure for the sensory systems. We then send this information to a 3D printer where our soft sensors are made.”
The printer injects a conductive black paste into liquid silicone. The silicone hardens, but the paste has been enclosed by it and remains liquid. When the sensors are subjected to squeezing or stretching, there is a change in the electrical resistance.
“That tells us how much compression or stretching force is applied to a surface. We use this principle to gain a general understanding of interactions with objects and, specifically, to learn how to control an artificial hand interacting with these objects,” explains Hidalgo.
What sets their work apart: the sensors fixed in silicon adjust to the surface in question (such as fingers or hands) but still offer accurate data that could be utilized for the interaction with the surrounding.
New Perspectives for Robotics and Especially Prosthetics
The integration of these soft, skin-like sensors in 3D objects opens up new paths for advanced haptic sensing in artificial intelligence.
Sami Haddadin, Executive Director and Professor, Munich Institute of Robotics and Machine Intelligence, Technical University of Munich
The sensors offer useful data on compressive forces and deformations in real-time–thus offering immediate feedback. This extends the range of perception of an object or a robotic hand—streamlining a highly advanced and sensitive interaction.
This work has the potential to bring about a general revolution in industries such as robotics, prosthetics and the human/machine interaction by making it possible to create wireless and customizable sensor technology for arbitrary objects and machines.
Sami Haddadin, Executive Director and Professor, Munich Institute of Robotics and Machine Intelligence, Technical University of Munich
Journal Reference
Groß, S., et al. (2023) Soft Sensing Skins for Arbitrary Objects: An Automatic Framework. IEEE International Conference on Robotics and Automation. doi.org/10.1109/ICRA48891.2023.10161344.
Source: https://www.tum.de/en/