REGISTER NOW

Physics AI for Digital Twins: Cutting Test Cycles and Program Risk in Aerospace & Defense

When: June 30, 2026
Time: 10:00 am PT / 1:00 pm ET

image_prefix_0_(16)
Overview:

In Aerospace and Defense, a valuable digital twin of an aircraft or missile system has three key capabilities - predict the complex physical interactions between the various domains, learn and adapt from real world data continuously and react fast enough to provide actionable feedback.

For a missile system that typically operates in non-linear aerodynamic regimes, building an accurate digital twin that is multi-domain, adaptable and fast can be prohibitively complex.

Physics AI has the potential to address this challenge by predicting high-fidelity fields (and derived quantities of interest) in seconds.

In this webinar, we demonstrate a Digital Twin reference architecture using SHIFT-Missile, a  Physics AI model for missile aerodynamics and structural mechanics.

What You'll Learn:
  • How the model is built: geometry parameterization, large-scale data generation, and training approach
  • Model performance highlights: agreement with CFD across predicted quantities and what that enables for engineering workflows
  • Continuous Model Correction: to bridge the real world and virtual environment.
What This Enables:
  • Coupled workflows: Rapid aero-structural iteration, sensitivity analysis, uncertainty quantification, etc.
  • Real-time Simulator: Realistic simulator with high-fidelity Physics AI engine
  • Multidisciplinary Design Optimization: Embed inference directly inside optimization loops to evaluate thousands of candidates
  • Trajectory/mission analysis: Rapidly evaluate aero-structural performance to analyze system limits for safe and feasible maneuvers
Who Should Attend:
  • Aerospace & defense engineering teams evaluating faster ways to generate and operationalize aerodynamic insight
  • CFD simulation engineers, GNC Engineers, Loads and Dynamics, Structural engineers and Thermal Engineers interested in Physics AI surrogates
  • R&D and technical leadership exploring scalable model-driven design workflows
  • AI/ML practitioners working on geometry-conditioned learning, PDE surrogates, or physics-informed system

REGISTER NOW


Meet the speakers:

Andrew Hong

Andrew Hong

Luminary
Forward Deployed Engineer

Andrew holds a PhD in Mechanical Engineering for Purdue University, where he investigated rarefied gases and powders. Prior to joining Luminary, he was a postdoc at Sandia National Laboratories, where he pivoted towards modeling hypersonic applications, pore-scale physics in thermal protection systems, and particulates in rarefied environments.

Dheeraj Vemula-1

Dheeraj Vemula

Luminary
Technical Marketing Engineer

Dheeraj Vemula has 8 years of experience in the CAE simulation space. Dheeraj specializes in the of AI/ML-driven simulation and digital twins. His background spans product development and application engineering, underpinned by a Master’s in Mechanical Engineering from NCSU and a Bachelor’s from IIT Madras.