Developing Digital Twins

A good example of how models can be built from data to predict the behavior of systems.  Here a UAV as an example.

Developing a Digital Twin
Oden Institute for Computational Engineering and Sciences (University of Texas at Austin)
December 4, 2019

Researchers at the University of Texas at Austin's Oden Institute for Computational Engineering & Sciences, the Massachusetts Institute of Technology, Akselos, and Aurora Flight Sciences are developing a predictive digital twin for a custom-built unmanned aerial vehicle (UAV). The twin, described by Oden director Karen Willcox at the 2019 International Conference for High Performance Computing, Networking, Storage and Analysis (SC19), uses physics-based models that capture the details of its behavior to represent each component of the UAV, as well as its integrated whole. The twin also analyzes on-board sensor data from the UAV and integrates that with the model to create real-time predictions of the health of the vehicle. The team paired computational modeling with machine learning to produce predictions that are reliable and explainable. Learning from data "brings together the methods and the approaches from the fields of data science, machine learning, and computational science and engineering, and directs them at high-consequence applications," said Willcox.... '