Programmable Architecture

-Towards Human Interactive, Cybernetic Architecture-

Kensuke Hotta,  Architectural Association School of Architecture

プログラマブル アーキテクチャ


堀田憲祐, 英国建築協会建築学校 

Chapter 6
Experiment 1: Simple Real-Time GA with Grasshopper


6-1. Introduction

Continuing the focus on illumination in Chapter 5, this chapter looks at a series of experiments to measure the shading capacity of various roof types under different light exposures using Rhinoceros and Grasshoppers. The special feature of this construction model was that, unlike a traditional CAD model, it was not static but dynamic using a variety of techniques discussed later. The construction model was made with 'Kangaroo Physics', a Grasshopper plug-in which contains a physics engine allowing physical modelling, and 'Galapagos', another plug-in which contains the GA (genetic algorithm) components and was used for optimization. Real-time experiments require a dynamic model as well as a set time period. This second aspect, the set time period, became a separate research issue.

6-1 . 序論


Fig.6-1,1: Program Execution Display ShotThis figure shows the interfaces of Rhinoceros and Grasshopper and Galapagos

6-2. Aim

The aim of these experiments was to prove that the proposed kinetic roof adapts most effectively to sun exposure over time providing the best result. In the first experiment, the number of shadows was counted by using the ‘exposure’ component from Grasshopper-the greater the area of shade, the better. In other words, the Objective Function (the necessary light level) was a constant in this experiment. The second experiment measured how much three different roofs systems-fixed, pre-optimized and kinetic-could follow a fluctuating Objective Function. This second experiment measured the system's adaptability, focusing on two different issues - the margin of error in trying to match to Objective Function and the other the speed at which they adapted.

6-2 . 目的


Continue on the book, kindle~