Fully Automatic Transfer And Measurement System For Structural Superlubric Materials

This article discusses the development of a fully automated system that combines deep learning and image recognition techniques to rapidly select, transfer, and measure tribological properties of SSL materials. The system is capable of efficient transfer and measurement on graphite flakes with different sizes and covers, and the selection accuracy and transfer success rate exceed 98% and 94%, respectively. The system is also able to perform comprehensive friction performance tests on more than 100 selected flakes.