| Abstract Scope |
A central challenge in automating and advancing complex welding manufacturing processes is that key quality metrics are often not directly measurable in real time. This talk presents a perspective on why deep learning has become a game changer in addressing this challenge. However, whether the measured process signals contain sufficient raw information to predict these hidden outputs is not easily determined. To address this issue, the University of Kentucky introduced the use of generative AI to test information adequacy. This talk will present further examples, including predictive human control in human–robot collaborative welding processes, and demonstrate how this approach offers a new foundation for building smarter, more adaptive welding systems through the prediction of complex welding phenomena. |