The Evolution of Conceptual Building Architectures

by Corrado Coia (year7c0 AT gmail . com)
Supervisor: Brian Ross

An evolutionary approach to the automatic generation of 3D building topologies is presented. Genetic programming is used to evolve split grammars, which are a refined kind of shape grammars. The grammars are interpreted by the CityEngine product, which is used for 3D building model generation. Fitness evaluation considers user-specified criteria that evaluate different aspects of the model geometry. Such criteria might include maximizing the number of unique normals, satisfying target height requirements, and conforming to supplied shape contours. Multi-objective evaluation is used to analyze and rank model fitness, based on the varied user-supplied criteria. A number of interesting models complying to given geometric specifications have been successfully evolved with the approach. A motivation for this research application is that it can be used as a generator of conceptual designs, to be used as inspirations for refinement or further exploration.

Publications

MSc thesis: "Automatic Evolution of Conceptual Building Architectures".
Corrado Coia. Brock COSC TR CS-1106, May 2011. (PDF)
"Automatic Evolution of Conceptual Building Architectures".
C. Coia and B.J. Ross. CEC 2011, New Orleans, June 2011. Slides (pptx)
Tech report: "Automatic Evolution of Conceptual Building Architectures".
C. Coia and B.J. Ross. Brock COSC TR CS-11-02, January 2011. (PDF)
GP evolves the shape grammar, and evaluates the suitability of the resulting models.
The GP system used is RobGP.
Shape grammars are interpreted by CityEngine, which then generates the resulting 3D model files.
CityEngine is a product of Procedural, Inc.

Examples

Image gallery.
Animations:
- Blue: (WMV)
- Angel Tower: (WMV)
- Space craft: WMV

Back up: http://www.cosc.brocku.ca/~bross/