Shape Optimization using Geodesic Convolutional Networks

ICML 2018

Our ICML paper about our approach to Aerodynamic shape optimization. This new contribution using the Geometric Deep Learning methods, improves upon previously used Kriging-based surrogate models for CFD optimization.

ICML 2018 Paper.

 

Online Pressure and Drag prediction with GCNN.

Optimization Results

Optimization of a foil with control points.

Finding the most aerodynamic shape around a sphere.