Paper Info
Title
Evolutionary swarm design of architectural idea models
Abstract
In this paper we present a swarm grammar system that makes use of bio-inspired mechanisms of reproduction, communication and construction in order to build three-dimensional structures. Ultimately, the created structures serve as idea models that lend themselves to inspirations for architectural designs. Appealing design requires structural complexity. In order to computationally evolve swarm grammar configurations that yield interesting architectural models, we observe their productivity, coordination, efficiency, and their unfolding diversity during the simulations. In particular, we measure the numbers of collaborators in each swarm individual's neighborhood, and we count the types of expressed swarm agents and built construction elements. At the end of the simulation the computation time is saved and the created structures are rated with respect to their approximation of pre-defined shapes. These ratings are incorporated into the fitness function of a genetic algorithm. We show that the conducted measurements are useful to direct an evolutionary search towards interesting yet well-constrained architectural idea models.
Year
DOI
Venue
2008
10.1145/1389095.1389115
GECCO
Keywords
Field
DocType
swarm grammar system,evolutionary swarm design,interesting architectural model,architectural idea model,swarm grammar configuration,construction element,appealing design,architectural design,swarm agent,idea model,swarm individual,fitness function,boids,complexity,genetic algorithm,structural complexity
Mathematical optimization,Swarm behaviour,Computer science,Boids,Fitness function,Multi-swarm optimization,Artificial intelligence,Stigmergy,Machine learning,Genetic algorithm,Computation,Swarm robotics
Conference
Citations 
PageRank 
References 
14
1.00
15
Authors
2
Name
Order
Citations
PageRank
Sebastian Von Mammen112624.68
Christian Jacob221133.00