A robotic sorting system developed by researchers at Tsinghua University achieved a 98.85% accuracy rate in recognizing various domestic waste items by mimicking the human sense of touch.
This state-of-the-art technology could improve recycling efficiency and potentially aid people with hand disabilities. The system uses tactile sensing, thermal sensations, and machine learning algorithms to identify and classify objects accurately. The team designed a layered sensor capable of detecting thermal conductivity, surface roughness, contact pressure, and temperature. When tested, the robot sorted common trash items into appropriate categories with high precision.
Future research may involve integrating this sensor with brain-computer interface technology to restore tactile perception in individuals with hand disabilities.