This research direction explores how artificial intelligence can transform conventional materials research.

To address the high dimensionality of growth parameters, long experimental cycles, and strong dependence on empirical trial-and-error in two-dimensional materials research, we integrate materials experimentation, laboratory automation, robotics, machine vision, in-situ characterization, and artificial intelligence into an autonomous experimental platform.

Through automated experimentation, real-time acquisition of multimodal data, machine-learning modeling, and closed-loop optimization, the experimental system can learn from previous results, recommend new experimental conditions, and automatically execute subsequent experiments. Our goal is to transform materials research from conventional trial-and-error approaches toward high-throughput, automated, data-driven, and increasingly autonomous experimentation.

Research Topics:

· Automated CVD and robotic experimental systems

· Multimodal in-situ characterization and machine vision

· Automated wafer-scale Raman, optical, and electrical characterization

· Materials experiment databases and machine-learning models

· Bayesian optimization, active learning, and reinforcement learning

· AI-driven experimental design and closed-loop autonomous optimization

Long-Term Goal: To develop a Self-Driving Materials Laboratory for two-dimensional materials research.