EXTENDED ABSTRACT: Along with a worldwide increasing popularity of deep learning in computer science that began in the 2010s, it has been actively applied in the field of mechanical engineering and materials science as well, such as in processing, optimization, manufacturing, experimental fluid dynamics, finite element method (FEM) and computational fluid dynamics (CFD) simulations, topological design, and robotics. Unlike ordinary numerical method of solving governing differential equations (physics-based), deep learning has presented a completely new perspective of analysing modern technologies. Trained from a given dataset (experiment or simulation), the deep learning model can predict the future with very good accuracy, by spontaneously discovering intrinsic patterns contained in the data (thus, data-driven). In addition, very fast predictions such as real-time analytics are possible. In this sense, I would like to present ‘Innovative applications of artificial intelligence (AI) to mechanical/materials processing’, based on research experiences in deep learning applications to core technologies for automobiles, vessels, drones, and manufacturing machines.
Keywords: Deep learning, Computer vision, Mechanical processing, Materials processing