Improving Wafer Displacement Algorithms Using Generative AI for Synthetic Training Image Augmentation

Modern semiconductor manufacturing demands extremely high wafer positioning accuracy to prevent defects such as interlayer misalignments, bridging, and incomplete etching, which degrade device performance and impact production yield. Although feedback-controlled robotic systems achieve high positional precision, placement errors may still arise due to surface non-uniformity, thickness variations, and mechanical drift. To address these challenges, this study applies generative artificial intelligence (AI) techniques to augment wafer image datasets, aiming to enhance algorithm robustness, improve positioning sensing accuracy, and strengthen the operational stability of automated systems. The proposed approach demonstrates strong potential for advancing intelligent wafer alignment technologies and establishes a solid foundation for enhancing the reliability and scalability of next-generation semiconductor manufacturing.