Reproduce “A Probabilistic Graphical Model Foundation for Enabling Predictive Digital Twins at Scale” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Michael G. Kapteyn, Jacob V.R. Pretorius, Karen E. Willcox
A formal mathematical framework for digital twins is established through a probabilistic graphical model, a Dynamic Decision Network, designed to enable scalable implementations with integrated uncertainty quantification. The approach successfully calibrated a UAV structural digital twin and demonstrated adaptive in-flight health monitoring, which adjusted mission maneuvers based on estimated structural degradation.
Key insights
- Probabilistic graphical models offer a robust and flexible mathematical foundation for designing scalable and rigorous digital twins by formally encoding system dependencies.
- End-to-end uncertainty quantification is essential for reliable digital twin performance, allowing for principled estimation, prediction, and adaptive decision-making under real-world variability.
- Digital twins, when built upon such a framework, can evolve from static monitoring tools to adaptive, decision-making systems capable of optimizing physical asset operations and ensuring safety.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.