Structural Design with Machine Learning
The Combinatorial Equilibrium Modelling (CEM) is a form-finding tool for structural design based on 3D graphic statics and graph theory that
allows for the
interactive generation of mixed tension-compression equilibrium shapes across different non-typological design spaces.The exploration of the
design spaces associated to complex three-dimensional structures using the CEM, however, might prove a difficult task due to the large amount
of parameters that have to be controlled by the designer.
In this research project, the potential to combine the CEM with machine learning is presented as part of an ongoing research collaboration between the Chair of Structural Design and the Chair for Computer Aided Architectural Design (CAAD) at ETH Zurich. A design framework is developed that takes advantage of the interaction between human and machine intelligence. Thanks to algorithms such as Self Organizing Map (SOM) and Gradient Boosting Trees, in the proposed approach the machine is used to guide and support the designer in the quest for innovative and appropriate solutions for structural design tasks that go beyond the conventional structural typologies.
Ole Patrick Ohlbrock, Pierluigi D′Acunto (ETH Zurich, Chair of Structural Design)
Vahid Moosavi, Karla Saldana (ETH Zurich, Chair for Computer Aided Architectural Design)
Lukas Fuhrimann, Vahid Moosavi, Patrick Ole Ohlbrock, Pierluigi D'Acunto: Data-Driven Design: Exploring new Structural Forms using Machine Learning and Graphic Statics, Proceedings of the IASS Symposium 2018 - Creativity in Structural Design, Boston, 2018.
last modified 23.9.2019