Slide 01 - Cover Page
The goals of this project were not to design a building but to Develop a method of working.
Slide 02 - The criteria were that it be:
Flexible: for a variety of programs and sites
Self-organizing at a variety of scales, using simple rules
“get away from the surface” meaning: parts interrelate in 3D space
Incorporate both topological and geometrical response to stimuli
Slide 03 - RESEARCH
- I began looking at biological systems for examples of phenomena that I could translate to a digital process
Slide 04 - I began with cell differentiation, wanting to find the mechanism by which cells become different organ assemblies, for example bone, or blood cells each with very different types of structures and functions.
Slide05 - This is the body plan of the fruit fly (drosophilia) during the first 10 hours of life, showing the development of different organs, all from a single cell.
Slide06 - As we look closer, the driving forces in the variation and assembly of these body systems are the cell transcription factors: coded molecules that control where and when chains of genetic material are essentially cut and pasted together to form new cellular material. Transcription factors are triggered by environmental qualities, such as temperature and concentrations of ambient chemicals.
Slide07 -When Triggered, transcription factors sometimes do not act directly on the genetic material of the cell, but on other transcription factors, combining action and reaction in complex chains of causality. Here, the boundary between organism and environment is blurred, each constantly influencing the other.
Slide08 - As I looked at other natural systems, the same phenomena kept popping up: the physical presence of certain chemicals in the environment conveyed very specific information for organisms at the cellular level.
Slide09 - The cells of a slime mold colony can exist in multiple states, they can roam free as single-cell organisms, but when collected together as a larger mass, they develop organs and a fruiting body to reproduce. In order to navigate these different states they communicate with each other using pheromones.
Slide10 - I also studied “chemotaxis” – the physical motion of cellular organisms in direct response to chemical cues – this video shows a human white blood cell’s locomotion in response to chemicals released into its environment.
Slide11 - But I eventually came back to cell transcription, in a simplified form. After all, it’s hard not to identify the systems in our bodies with those in a building, and morphogenetic processes with design and assembly. This initial diagram simply maps all of the components of the cell transcription process onto a script-based design process.
Slide12 - APPLICATION
Slide13 - To develop a script-based process, involving the constant feedback between organism and environment, that was flexible enough to apply in a variety of ways, I had to simplify the initial diagram. There are two major elements in the system: the “cells” represented by groups of geometry, and pheromones represented by 3D points. Each instance of these elements has a specific name, consisting of the element’s location in space, how many generations its been around for, and its type (which I’ll get to later). The “cells” also have a radius, which is used for packing and preventing overlap.
Slide14 - Here are early example of a 2D process in which the cells are simply round shapes who’s size roughly corresponds to their radii. In these studies, the pheromones are broadcast at random by each cell as It reproduces. Cells reproduce within a line that represents the site boundary of a project. They either change their type or die off, depending on the concentration of pheromones within range.
Slide15 - More recent development of the script involves the three dimensional location of surrounding geometry, using a coded array of 26 directions about a given point. For example, in the diagram on the right, lines projected from the red dot to the points around it pass through the highlighted regions of the surrounding framework. In order to act on each cell in response to its environment, the script assigns numbers to the points, which can then be used to assess relative location and make attachments.
Slide16 - Here you see the script in operation with a simple space frame component used as the cell. In this case, the site boundary has been extruded to represent the zoning height limit, forming a 3D volume. When the cell detects proximity to the boundary, it emits a pheromone. Also, if it detects pheromones within its range, it will emit more. Local concentration of pheromones causes a change in the parameter that controls the thickness of the struts. As concentration reaches a critical threshold, the cell type changes. The idea here is that the system can accommodate multiple cell types that are completely different in size, geometry and topology, but all share common connectivity.
Slide17 - The recursive process of acting on each cell in the population is divided into three parts, loosely based on the 7 qualities that define living organisms. (homeostasis, organization, metabolism, growth, response to stimuli, reproduction). The first phase consists of gathering information from the surrounding environment, the second of releasing information into the environment, and the third of adapting and reproducing based on the first two phases.
Slide18 - This is an example of the same set of rules, but a change in the density of pheromones that cells need to emit new pheromones. In this case, the new cell type begins to appear at the site boundary, but soon spreads through the rest of the system.
Slide19 - In this final example, cells respond to two pheromone types. The first represents the boundary condition. The other, an interior condition: type 2 pheromones are released when cells detect other cells on more than two sides. With sufficient concentration of the type two pheromones, the cells go into a “dormant” state: still present in the awareness of other cells, but invisible to us, representing open space in side the building.
Slide20 - ... And a rendering of the same script run in a different environment. In Conclusion, I would like to say that this is still very much a work in progress. I am very open to suggestions brainstorming, and even collaborative adaptation of this work in order to deepen and improve the process.