Honolulu, Hawaii
June 24, 2007
June 24, 2007
June 27, 2007
2153-5965
Architectural
7
12.320.1 - 12.320.7
10.18260/1-2--2004
https://peer.asee.org/2004
451
D. Hoffers is student at Delft University of Technology. He participated 2006 in the BuildingRelations project and developed the sub-tool: SizeDefiner.
M. Frederiks is student at Delft University of Technology. He participated 2006 in the BuildingRelations project and developed the sub-tool: FunctionDistributor.
S. Korebritz is student at Delft University of Technology. He participated 2006 in the BuildingRelations project and developed the sub-tool: BoundingBox.
BR: An Interactive Software-Prototype for 3D Layout
Abstract
As a research project implemented with graduate students from TU_____, BuildingRelations [BR] deals with the development of an interactive software prototype to support the design process: BR employs bottom-up principles of organization to generate functional layouts exhaustively enabling development of more alterna- tives than by means of conventional sketching methods mainly because architectural space planning is highly combinatorial, and therefore, difficult to conceive exhaustively by human search means.
Content
Focusing on the development of an interactive design tool which allows simulation of complex design proc- esses, the project proposes an alternative design method based on Swarm Intelligence [SI]. SI is, basically, an Artificial Intelligence [AI] method consisting of agents interacting locally with one another and with their en- vironment similarly to the way fish interact in a swarm and birds in a flock.
In the absence of top-down control dictating how individual agents should behave, local interactions between agents lead to the bottom-up emergence of global behavior. The rules according to which agents interact are simple: C. Reynolds' flocking simulation, for instance, is based on three rules according to which digital birds flock – [1] maintain a minimum distance from neighbors, [2] match velocity with neighbors and [3] move to- wards the center of the swarm. While these rules are local, establishing the behavior of one agent in relation- ship to its neighbor, the flock behaves as a whole coherently.
Figure 1: Functional units swarm in the 3D space towards local optimal configurations. Functional units are represented in this example as blue boxes; spatial relations between functional units are represented as magenta connecting-lines.
Similarly, all functional units/objects pertaining to a building can be seen as flocking agents striving to achieve an optimal layout. Spatial relations between functional units can be described as sets of rules, according to which all units organize themselves into specific configurations. This approach is particularly suitable for the functional layouting of large and complex structures: While the architect might find it difficult to have an over- view of all functions and their attributed volume and preferential location, the functional units can easily swarm towards local optimal configurations.
Bier, H., & Hoffers, D., & Frederiks, M., & Korebritz, S. (2007, June), Br: An Interactive Software Prototype For 3 D Layout Paper presented at 2007 Annual Conference & Exposition, Honolulu, Hawaii. 10.18260/1-2--2004
ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2007 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015