PATTERN-Skin: Modular Multi-Modal Proximity and Tactile Perception Skin
Duration:
05/2021
—
03/2023
Total Time:
2 Jahre
The Project
Can robots feel?
In the project PATTERN-Skin, we developed an innovative, bendable and stretchable robot skin. This skin enables comprehensive perception of the environment for safe interaction between humans and robots. Realistic real-time simulations of the sensor skin allow optimal adaptation to different robots and applications. The simulations are also used to generate training data for machine learning.
Our activities within the project
In collaboration with Alpen-Adria-University Klagenfurt, we have developed a novel, bendable and potentially stretchable multimodal modular robot skin. This sensor skin enables us to equip robots with a variety of sensory capabilities and thus ensure even safer interaction between humans and robots.
Research group
Downloads
Sponsors
EFRE
KWF
REACT-EU
Project Partners
Alpen-Adria-Universität Klagenfurt
Fachhochschule Kärnten
Silicon Austria Labs GmbH
Project Details
No matter how innovative today's robots are, their ability to physically perceive their surroundings is limited compared to humans. Humans can rely on the receptors on their skin and sense the properties and textures of different materials. The PATTERN-Skin project aimed to address precisely this issue: giving robots a sense of touch. The lack of such integrated sensors in production means that robots have to carry out their work behind barriers, separated from humans. Robots are frequently isolated and used for simple, slow and suboptimal tasks to avoid unwanted physical interactions. However, this means that in many areas, robots do not fulfil user requirements and do not offer any added value. There is an urgent need to equip robots with capabilities to interact safely and effectively with humans and their environment.
Project aim
The project PATTERN-Skin aims to develop an innovative, flexible and potentially stretchable modular robotic skin and integrate it physically and logically into robots to give them unprecedented sensory capabilities. The use of non-contact and tactile perception enables safe interaction between humans and robots. In addition to the physical realisation of the robot skin, it is also crucial to develop realistic real-time simulations of the skin in simulations (digital twin). These simulations facilitate optimal adaptation of the sensor skin to different robots and applications They are also used to generate training data for machine learning and artificial intelligence in order to develop safe control strategies that meet the requirements of the ISO 15066 and 10218 safety standards.
