Can a stiff robot feel the softness of pillows, stuffed toys, and cotton candy like humans do? In reality, enabling robots to understand and reproduce the sensation of softness is extremely challenging, involving various complex sensory and cognitive processes. Swiss scientists have developed a robot softness perception interface that allows robots to sense the softness of real materials. The related findings were published on March 11th in the Proceedings of the National Academy of Sciences of the United States of America.
The research team from the Swiss Federal Institute of Technology in Lausanne has developed a robot interface called SORI (Softness Rendering Interface) that can distinguish between tactile and kinesthetic cues, allowing the robot to perceive the softness of various materials. The researchers first developed geometric parameters for the fingertips and their contact surfaces to estimate the softness of the fingertips. They then extracted softness parameters from a range of different materials and input them into the SORI device.
The researchers conducted experiments using a paper-folding robot equipped with motor-driven joints and silicone membranes on the top of the joints, which can expand or contract with airflow to adjust the softness. By simultaneously perceiving tactile and kinesthetic cues, SORI successfully reproduced the softness of materials such as beef, salmon, and marshmallows in multiple experiments. It also imitated complex materials like marshmallows covered with cookies and books bound in leather. In a virtual experiment, SORI even reproduced the sensation of a beating heart, demonstrating its ability to perceive softness during motion.
According to Mustafa Mete, a doctoral student from the Reconfigurable Robotics Laboratory at the Swiss Federal Institute of Technology in Lausanne, the goal of this research is not to serve as a softness sensor for robots but to digitally convey the sense of touch. The research team believes that this study fills a gap in robot research and opens up many applications for robots to utilize softness perception, such as deep-sea exploration, robot-assisted surgery, malignant tumor detection, and vegetable and fruit harvesting.
For more information, you can refer to the related paper: https://doi.org/10.1073/pnas.231490112
