Robots don’t have human anatomy to let them know about hardware or a software issue. If a part of its artificial body malfunctions then the robot will think that that is its normal function. To overcome this hurdle, scientists need skins or neural-like sensors to communicate this information three-dimensionally and continuously to the robot’s controller.
Researchers have developed a foam based on an artificial intelligence system that allows the robot to sense the movement of the foam. Silicone made key sensors with 30 optical fiber layer sticking out of one end of the foam connecting to other equipment. The light coming out from the end of the optical fibers represent the movement of the foam. The light looks a certain way when it is at rest, and it changes when the foam moves. The overall pattern of light intensity can help detect the changes in shape.
Robots lack human intelligence, so the researchers turned to artificial intelligence. To build the AI, the researchers first collected the data on how the light changed when the foam was twisted in a certain way. After obtaining the data, the scientists used this data to train the machine learning models that can further be used to interpret foam activity. However, lights are not the only sensing strategy that researchers use. Flexible electronic sensors utilize a variation in current to notice how they’re stretching. Stretchable light fibers have been used in the past to measure deformity.
The researchers admit that the foam and AI setup are experimental and are external to the foam but in the future, it would be miniaturized with the goal of a self-contained, self-sensing foam set up. Scientists believe that robots must learn to walk themselves and they must have the ability to sense their shape. Smart foam and artificial intelligence will assist robots to identify their injuries. In a few years, soft robots could be seen everywhere that might lead search and rescue missions and even administer medication to the injured.