Skip to content

Grand Challenge #5

Predict Internal Building Structures with Machine Learning

Out of a field of twenty-four white paper submissions and two letters of support, Karagozian & Case, Inc. (K&C) was awarded the Air Force Research Laboratory (AFRL) Grand Challenge 5. Grand Challenge 5 was created to provide a solution in predicting the internal structure of a building using machine learning.  The overarching goal is to predict the internal configuration of structures such as buildings by only using external satellite photographs taken from various angles.

K&C presented their solution, ASPEN (Adversarial Structural Predictor and Extraction Network), to the Air Force Research Laboratory (AFRL) during the November 12, 2021, Pitch Day as a solution. Using the shape, size, and external characteristics of the building, the desired methodology should be able to produce a model of the building to include a basic structure system. K&C’s machine learning model trained with a sufficient database of known buildings, along with constraints to represent current construction processes, could potentially yield a machine learning model that can use minimal external visual clues and then automatically produce a digital file of the most likely structural design of the building.

K&C is headquartered in Glendale, California, and is an internationally recognized science and engineering consulting firm founded in 1945. They have been a leader in protective design, analysis, and engineering for over five decades. Their mission is focused on solving complex and challenging problems using state-of-the-art testing, analytic, and design methodologies developed by the company. The company is excited for this opportunity to use their combined expertise in Machine Learning and Building Structures to develop this new technology for AFRL.

This challenge was awarded through the Wright Brothers Institute, Doolittle Institute, and the National Security Innovation Network.

TELL SOMEONE ABOUT THIS
Share on facebook
Share
Share on twitter
Tweet
Share on linkedin
Share