Swarm Intelligence in Architectural Design

Home , Bait ball, Bat algorithm, Boids, Feeding frenzy, Flocking (behavior), Herd

Yuxing Chen

Advisor: Ronald Rael Raveevarn Choksombatchai

Content

Chapter 1: Introduction of swarm 1.1 Thesis Statement 2 1.2 Swarm Behavior 3 1.3 Mathematical Models 6 1.4 Boids System 7 1.5 Swarm Intelligence 9

Chapter 2: Swarm algorithms and application 2.1 Swarm Intelligence in Stadium design 12 2.2 Swarm used at Design (by Tyler Julian Johnson) 14 2.3 Swarm Intelligence at Visualization (Robert Hodgin) 18 2.4 Complexity of Swarm at Arts 22 2.5 Object to Field 25 2.6 Swarm Tectonics 29 2.6 Swarm Urbanism (Neil Leach) 31 2.7 Swarm Modeling 33 2.8 Motion at Architecture Design 34 2.9 Particles at Architecture Design 35

Chapter 3: Swarm Tectonics 3.1 Swarm Testing 42 3.2 From Simulation to Application 44 3.3 Swarm Structure 51 3.4 Skin Attachment 60 3.5 Swarm Modular 63 3.6 Swarm Joint 68 3.7 Swarm Drawing 74 3.8 Data Swarm 90 3.9 Conclusion 97 Bibliography 98

Chapter1: Introduction

1 1.1 Thesis statement

Swarm behavior is a collective behavior exhibited by animals of similar size which aggregate together. It can be applied to any other animal that exhibits swarm behavior.

Swarm intelligence is the collective behavior of decentralized, self-organized systems, natural or artificial. The basic idea of swarm intelligence is a “population” of local interactions to the environment in a greater amount that creates a global system.

At 1989, the swarm intelligence expression was first introduced at robotic systems, which describe the emergent collective behavior. Nowadays this new approach of collective behavior has diversely implemented in many perspective from biology, social structure, engineering, artificial, visualization and architecture. What’s more, a collection of people can also exhibit swarm behavior, such as pedestrians.

From the book, “Emergence”, by Steven Johnson, he also wrote about creating a “form” of living on having emergence logic from the smallest scale as ants into a larger scale as cities. He tries to define how the swarm logic of ants’ behavior could give example of the way of living for human level.

For architects, swarm intelligence examines the role of agency within generative design processes. Digital parametric programs like processing are used to make high populations of self-organized elements into an emergent intelligence. The gold is to explore the dissolution of modernist tectonic hierarchies. The research about application of swarm principles at architecture ranges from visualization, self-organization of multi-agent system, architecture form design, urbanism, etc.

Based on the boids system at the digital animation tool, we can simulate the swarm activity, translate the animal particle system with architecture tectonic system and research about architecture form, structure and space. Swarm could be generated into an architecture design method.

2 1.2 Swarm Behavior (http://en.wikipedia.org/wiki/Swarm_behaviour)

As a term, swarming is applied particularly to insects, but can also be applied to any other animal that exhibits swarm behaviour. The term flocking is usually used to refer specifically to swarm behaviour in birds, herding to refer to swarm behaviour in quadrupeds, shoaling or schooling to refer to swarm behaviour in fish.

BY BRENDAN SEIBEL

BIRD FLOCKING

3 BY OCTAVIO ABURTO

FISH SCHOOLING

4 FROM MAASAI MARA KENYA NATIONAL RESERVE

BUFFALO HERDING

5 SWARM INTELLIGENCE IN ARCHITECTURE DESIGN

1.3 Mathematical models (http://en.wikipedia.org/wiki/Swarm_behaviour) SWARM behaviour, or swarming, is a collective behaviour exhibited by animals of similar size which aggregate together, perhaps milling Early studies of swarmabout behavior the same spotemployed or perhaps mathematical moving en masse or models migrating into some simulate direction. As a term, swarming is applied particularly to insects, and understand the behavior.but can also Thebe applied simplest to any other mathematical animal that exhibits models swarm of behaviour. animal The term flocking is usually used to refer specifically to swarm behaviour in birds, herding to refer to swarm behaviour in quadrupeds, shoaling or schooling to refer to swarm behaviour in fish. swarms generally represent individual animals as following three rules: 1. Separation: Move in the same direction as your neighbors SWARM INTELLIGENCE2. Alignment: IN Remain ARCHITECTURE close to your neighborsDESIGN 3. Cohesion: AvoidBOIDS collisions is an artificial with your life program, neighbors developed by Craig Reynolds in 1986, which simulates the flocking behaviour of birds. His paper on this topic was published in 1987 in the proceedings of the ACM SIGGRAPH conference.The name “boid” corresponds to a shortened version of “bird-oid object”, which refers to a bird-like object

SWARM behaviour, or swarming, is a collective behaviour exhibited by animals of similar size which aggregate together, perhaps milling about the same spot or perhaps moving en masse or migrating in some direction. As a term, swarming is applied particularly to insects, but can also be applied to any other animal that exhibits swarm behaviour. The term flocking is usually used to refer specifically to swarm behaviour in birds, herding to refer to swarm behaviour in quadrupeds, shoaling or schooling to refer to swarm behaviour in fish.

SWARM INTELLIGENCE IN ARCHITECTUREBOIDS is an artificial life DESIGN program, developed by Craig Reynolds in 1986, which simulates the flocking behaviour of birds. His paper on this topic was published in 1987 in the proceedings of the ACM SIGGRAPH conference.The name “boid” corresponds to a shortened version of “bird-oid object”, which refers to a bird-likeSEPARATION object ALIGNMENT COHESION Steer to avoid crowding Steer towards the av- Steer to move toward the local flockmates erage heading of local average position (center flockmates of mass) of local flockmates SWARM behaviour, or swarming, is a collective behaviour exhibited by animals of similar size which aggregate together, perhaps milling about the same spot or perhaps moving en masse or migrating in some direction. As a term, swarming is applied particularly to insects, but can also be applied to any other animal that exhibits swarm behaviour. The term flocking is usually used to refer specifically to swarm behaviour in birds, herding to refer to swarm behaviour in quadrupeds, shoaling or schoolingSWARM to refer INTELLIGENCE to swarm behaviour (SI) in is fish. the collective behavior of decentralized, self-organized systems, natural or artificial. The concept is employed in work on artificial intelligence.

SI systems consist typically of a population of simple agents or boids interacting locally with one another and with their environment. The inspiration often comes from nature, especially biological systems. The agents follow very simple rules, and although there is no BOIDS is an artificial life program, developed by Craig Reynolds in 1986, which simulates centralizedthe flocking control behaviour structure of birds. dictating His paper how individual agents should behave, local, and to a certain degree random, interactions between on this topic was published in 1987 in the proceedingsSEPARATION of the ACM SIGGRAPH conference.TheALIGNMENTsuch name agents “boid” lead corresponds to the emergence to a shortened of “intelligent” COHESION global behavior, unknown to the individual agents. Examples in natural systems of version of “bird-oid object”, which refers to a bird-likeSteer to object avoid crowding SteerSI include towards ant the colonies, av- bird flocking, animal herding,Steer to move bacterial toward growth, the and fish schooling. The definition of swarm intelligence is still local flockmates eragenot quite heading clear. of In local principle, it should be a multi-agentaverage position system (center that has self-organized behaviour that shows some intelligent behaviour. flockmates of mass) of local flockmates

SWARM INTELLIGENCE (SI) is the collective behavior of decentralized, self-organized systems, natural or artificial. The concept is employed in work on artificial intelligence.

SI systems consist typically of a population of simple agents or boids interacting locally with one another and with their environment. The inspiration often comes from nature, especially biological systems. The agents follow very simple rules, and although there is no centralized control structure dictating how individual agents should behave, local, and to a certain degree random, interactions between such agents lead to the emergence of “intelligent” global behavior, unknown to the individual agents. Examples in natural systems of SEPARATION ALIGNMENTSI include ant colonies, bird flocking, animal herding,COHESION bacterial growth, and fish schooling. The definition of swarm intelligence is still Steer to avoid crowding Steernot quite towards clear. the In avprinciple,- it should be a multi-agentSteer to move system toward that the has self-organized behaviour that shows some intelligent behaviour. local flockmates erage heading of local average position (center flockmates of mass) of local flockmates 6

SWARM INTELLIGENCE (SI) is the collective behavior of decentralized, self-organized systems, natural or artificial. The concept is employed in work on artificial intelligence.

Boids is an artificial life program, developed by Craig Reynolds in 1986, which simulates the flocking behavior of birds. The name “boid” corresponds to a shortened version of “bird-oid object”, which refers to a bird-like object. Its pronunciation evokes that of “bird” in a stereotypical New York accent.

As with most artificial life simulations, Boids is an example of emergent behavior; that is, the complexity of Boids arises from the interaction of individual agents (the boids, in this case) adhering to a set of simple rules. The rules applied in the simplest Boids world are the three basic rules of swarm behavior: separation, alignment and cohesion.

At the paper, Flocks, Herds, and Schools: A Distributed Behavioral Model, it says that” the aggregate motion of a flock of birds, a herd of land animals, or a school of fish is a beautiful and familiar part of the natural world. But this type of complex motion is rarely seen in computer animation. This paper ex- plores an approach based on simulation as an alternative to scripting the paths of each bird individually. The simulated flock is an elaboration of a particle system, with the simulated birds being the particles. The aggregate motion of the simulated flock is created by a distributed behavioral model much like that at work in a natural flock; the birds choose their own course. Each simulated bird is implemented as an independent actor that navigates according to its local perception of the dynamic environment, the laws of simulated physics that rule its motion, and a set of behaviors programmed into it by the “animator.” The aggregate motion of the simulated flock is the result of the dense interaction of the relatively simple behaviors of the individual simulated birds.”

7 The model is based on simulating the behavior of each bird independently. Working independently. The birds try both to stick together and avoid collisions with one another and with other objects in their environment. The animations showing simulated flocks built from this model seem to correspond to the observer’s intuitive notion of what constitutes “flock-like motion. “The most interesting motion of a simulated flock comes from interaction with other objects in the environment. The isolated behavior of a flock tends to reach a steady state and becomes rather sterile. The flock can be seen as a relaxation solution to the constraints implied by its behaviors.

8 1.5 Swarm Intelligence ( http://en.wikipedia.org/wiki/Swarm_intelligence )

Swarm intelligence (SI) is the collective behavior of decentralized, self-organized systems, natural or artificial. The concept is employed in work on artificial intelligence. The expression was introduced by Gerardo Beni and Jing

SI systems consist typically of a population of simple agents or boids interacting locally with one another and with their environment. The inspiration often comes from nature, especially biological systems. The agents follow very simple rules, and although there is no centralized control structure dictating how individual agents should behave, local, and to a certain degree random, interactions between such agents lead to the emergence of “intelligent” global behavior, unknown to the individual agents. Examples in natural systems of SI include ant colonies, bird flocking, animal herding, bacterial growth, and fish schooling.

The definition of swarm intelligence is still not quite clear. In principle, it should be a multi-agent system that has self-organized behavior that shows some intelligent behavior.

The application of swarm principles to robots is called swarm robotics, while ‘swarm intelligence’ refers to the more general set of algorithms. ‘Swarm prediction’ has been used in the context of forecasting problems.

9 The examples of the algorithm include particle swarm optimization, ant colony optimization, artificial bee colony algorithm, differential evolution, the bee’s algorithm, artificial immune systems, grey wolf optimizer, bat algorithm, grav- itational search algorithm, altruism algorithm, glowworm swarm optimization, river formation dynamics, self-propelled particles, stochastic diffusion search and multi-swarm optimization.

10 Chapter2: Swarm algorithms and application

11 2.1 Swarm Intelligence in Stadium Design (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3703855/)

People in large-scale public areas are in danger because of a lot of manmade or natural accidents, such as fire, hurricane, and bomb. For coping with these emergencies, many scientists and engineers have paid much attention to the researches about evacuation routes planning. In these researches, the application of swarm intelligence technologies to evacuation routes planning is a hot topic because evacuation process itself is a collective behavior. Swarm intelligence technology mainly includes particle swarm optimization (PSO) technology and ant colony optimization (ACO) technology. The swarm intelligence technology is mainly used in two aspects: the simulation of evacuation process and the optimization of evacuation routes. On one hand, swarm intelligence technologies have natural advantages to simulate collective behavior such as evacuation process. On the other hand, the optimization mechanism of swarm intelligence algorithms can effectively optimize evacuation objectives by iterating the configuration of factors that affect evacuation efficiency. The factors that affect evacuation efficiency includes pheromone, location of shel- ters in evacuation zone, the direction of lanes, the placement of road barriers, and the scheduling of evacuation for each evacuee

12 Besides, evacuation routes optimization problem usually needs to consider multiple objectives, such as total clearance time total number of survivals. A few researches have involved the multi-objective evacuation routing optimization problem. Some of them applied swarm intelligence technologies to solve this kind of problem

13 2.2 Swarm used at Design (by Tyler Julian Johnson) (http://www.tyler-johnson.com/Swarm-Intelligence) At the project by Tyler Julian Johnson, the role of agency within a generative design process by using computational methodologies grounded in swarm intelligence and casting a simple decision making ability into agents capable of self-organizing into an emergent intelligence. The projects focused on technical code. Technical code writing required to cast swarm systems and the architectural theory behind these systems. The project developed simulations of vector based swarm systems and then used these systems as a basis for developing an architectural design methodology which operates within a topological substrate. The second half of the project (moving from a two-dimensional environment to a three-dimensional environment) shifted away from simply mapping agents over time and instead became a system capable of negotiating architectural inputs.

Computational design is shifting away from heavy systems (like Maya’s MEL scripting language) and into lighter weight object-oriented languages like Processing and Rhinoscript. This entire simulation was written in Processing, subdivided with a Rhinoscript, isosurfaced with Processing, and rendered with V-Ray for Rhino.

14 15 16 17 2.3 Swarm Intelligence at Visualization (Robert Hodgin) (http://roberthodgin.com/about/) Robert Hodgin’s work ranges from simple 2D data visualizations to immersive 3D terrain simulations about swarm behavior. His primary interests include theoretical physics, astronomy, particle engines, and audio visualizations. I work in C++, Cinder, OpenGL, and GLSL.

Massive anchovy swarms off the coast of California have kept marine mam- mals and their observers busy for the past couple of months. It’s not so much that there are more anchovy than usual, it’s that there are more anchovy gathered in one place. According to this article, anchovy movement can be due to a number of factors – plentiful plankton, mild temperatures – and this year, the anchovy stars aligned over Monterey Bay. Their presence, telegraphed far and wide via whale song, has set off a feeding frenzy of seals, whales, dolphins, and the press.

At the project, Magnetic Ink, it began as a tangent from the flocking studies. The thinking was simple. What if the flocking birds rained down a fine mist of ink onto a sheet of virtual paper? At the same time, they have ribbons that hang from their feet and if they fly low enough, the ribbon will drag on the paper and erase the ink.

18 The next project was flocking motion graphics for Nervo. What was required was a couple videos of flocking using a 3D crow (or is it a raven) they would provide. Simple enough. But given the tight deadline, the thought of doing a render and posting it and waiting for approval or changes and then implement- ing the changes then retendering and reposting, etc… That process didn’t make sense for this project so they decided to deliver them an application instead.

Using Processing, they started playing around with the flocking behavior to make it more customizable. The original version of the flocking experiment had very few controls and they had to be hard-coded. There was no run-time adjustment. This was the first thing addressed. Several new parameters were added. They included population density, gravity, drag, collision avoidance, flight range, camera position and tracking, and a few toggles such as tethering strings, floor plane, and bezier curves. Once the parameters were tweaked to the user’s liking, they need only to hit the spacebar and an image sequence of PNGs would start saving to the hard drive.

19 Once they had the exported image sequence, it was pretty easy to put it into a post processing application and work his magic.

At the bait ball project, they have been experimenting with simulating group dynamics for nearly as long as they have been coding. They became fascinated with the work of Craig Reynolds who was among the first to show that with very simple rulesets, predator and obstacle avoidance, and pursuit. Over a decade later, I watched a talk by Professor Iain Couzin who is the head of the Collective Animal Behavior department at Princeton. He explained the rules of flocking behavior in a way that really clicked. That was the impetus to make their first generative bait ball.

20 It was pretty easy to get these fish-objects to do something vaguely flock- like, but getting them to form into a torus proved to be a bit more of a chal- lenge. If you just throw a bunch of flocking objects into a 3D environment, often what happens is they come together into a clumped formation and then they just head off as a group and disappear into the GL fog. He needed a way to keep them corralled but without feeling like he was imposing unnatural restrictions on their behavior. Every time he have to add a new controlling vari- able, he feel he am moving away from the purity of the behavior. But he didn’t know what else to do so he introduced them to the concept of their own cen- troid. In addition to the flocking rules outlined by Reynolds, he added a general desire to not wander too far away from the average of all of their positions. This tiny change was all it required to get them to form stable and energetic toroid’s.

21 2.4 Complexity of Swarm at Arts (https://seeingcomplexity.wordpress.com)

Seems like swarm algorithms and flocking behavior are pretty popular these days. The work of some architecture students, Hyun Chang Cho, Jun Ho Cho, and Eun Ki Kang, all involved in some really cool visualizations.

22 Seems like swarm algorithms and flocking behavior are pretty popular these days. The work of some architecture students, Hyun Chang Cho, Jun Ho Cho, and Eun Ki Kang, all involved in some really cool visualizations.

23 There are more computational tools for swarmvis. An impressive tool (that you can actually use yourself now; paper here), it is used for modeling flocking behavior.

Originating in only a few simple rules, the program captures the idea that I have been discussing for the past few days regarding emergent properties of complex systems. As I argued in a previous post, what the human brain is particularly good at is recognizing patterns. Once we find patterns in complex systems, we can develop fairly simple rules that can explain these patterns (I also suggested that this is a radical departure from contemporary statistical methods, especially keeping in mind the prevalence of stochastic modelling). Don Miner and Niels Kasch (site here), authors of swarmvis, work backward to first form a set of simple rules that nonetheless produce quite complex results.

24 2.5 Object to Field

( Stan Allen, From Object to Field : Field Conditions in Architecture and Urbanism) Stan Allen’s ‘From object to Field’ clearly articulates an approach to heteroge- neous space in architecture and urbanism that contrasts both Modernist ideals of space as a uniform plane and Cubist concepts that informed Post-Modern- ist collage techniques. A series of historical case studies, such as Cordoba Mosque and non-architectural referents, offers a summary genealogy of the field within architecture and aesthetics in many ways, Allen’s use of the field complements Robin Evan’s idea of the ‘matrix’, extends the implications of Banham’s atmospheric architecture and can be understood as a translation of Deleuze and Guattari’s presentation of ‘smooth space’ into the realm of design. Allen’s presentation of ‘smooth space’ into the realm of design. Allen’s presentation does more than offer the ‘field’ as another design trope, in that field significantly alters the Modernist relationship between form, programme and space, as well as blurring the normative boundary between the discrete architectural building and larger urban forces and conditions. In that way, the article implies a deterritorialisation of disciplinary striations of the environ- mental disciplines, such as architecture, landscape and city planning, moving from the design of discrete artefacts to a choreography of multitudinous relations.

Field conditions move from the one toward the many, from individuals to col- lectives, from objects to fields. A distinct but related set of meanings begins with an intuition of a shift from object to field in recent theoretical and visual practices. In its most complex manifestation, ‘field conditions’ refers to mathematical field theory, to non-linear dynamics and computer simulations of evolutionary change. It parallels a shift in recent technologies from analog object to digital field. It pays close attention to precedents in visual art, from the abstract painting of 1920s to minimalist and post minimalist sculpture of the 1960s.

In the late 1980s, artificial life theorist Craig Reynolds created a computer program to simulate the flocking behavior of birds. Ad described by M Mitchell Waldrop in Complexity: the Emerging science at the Edge of Order and Chaos, Reynolds placed a large number of autonomous, bird-like agents, which he called ‘boids’, into an onscreen environment. The boids were programed to follow three simple rules which were introduced at the beginning of this booklet.

The flock is clearly a field phenomenon, defined by precise and simple local conditions, and relatively indifferent to overall form and extent. Because the rules are defined locally, obstructions are not catastrophic to the whole.

25 Faculty of Architecture, University of Sao Paolo, Brazil

Variations and obstacles in the environment are accommodated by fluid adjustment. A small flock and a large flock desply fundamentally the same structure. Over many interactions patterns emerge. Without repeating exactly, flock behavior tends toward roughly similar configurations, not as fixed type, but as cumulative result of localized behavior patterns.

Crowds present a different dynamic, motivated by more complex desires, and interacting in less predictable pattern. Elias Canetti in Crowds and Power has proposed a suggestive taxonomy: open and closed crowds; rhythmic and stag- nating crowds; the slow crowd and the quick crowd. He examines the varieties of the crowd, from religious throng formed by pilgrims, to the mass of par- ticipants in spectacle, even extending his thoughts to the following of rivers, the piling up of crops and the density of the forest. According to Canetti, the crowd has four primary attributes: the crowd always wants to grow; within a crowd there is equality; the crowd loves density; the crowd needs a direction. The relation to Reynolds’ rules outlined above is oblique, but visible. Canetti, however, is not interested in predictionor verification. His sources are literary, historical and personal. Moreover, he is always aware that the crowd can be liberating as well as confining, angry and destructive as well as joyous.

26 Photograph by Oto Bihalji-Merin

Iannis Xenakis, Syrmos, 1959

27 In attempting to reproduce what he referred to as ‘global acoustical events’, Xenakis drew upon his own considerable graphic imagination, and his training in descriptive geometry to invert conventional procedures of composition. That is to say, he began with a graphic notation describing the desired effect of ‘fields’ or ‘clouds’ of sound, and only later reduced these graphics to conventional musical notation. Working as he was with material that was beyond the order of magnitude of the available compositional techniques, he had to invent new procedures inorder to choreograph the ‘characteristic distribution of vast numbers of events’.

Crowds and swarm s operate at the edge of control. Aside from the suggestive formal possibilities, it was suggested with these two examples that architecture could profitably shift its attention from its traditional top-down forms of control and begin to investigate the possibilities of a more fluid, bottom-up approach. Field conditions offer a tentative opening in architecture to address the dynamics of use, behavior of crowds, swarm and the complex geometries of masses in motion.

28 2.6 Swarm Tectonics (Neil Leach , swarm tectonics: a manifesto for an emergent architecture)

In his book, Emergence: The Connected Lives of Ants, Cities and Software, Steven Johnson presents the city as a manifestation of emergence. The city operates as a dynamic, adaptive system, based on interactions with neighbors, informational feedback loops, pattern recognition and indirect control. ‘Like any emergent system’, notes Johnson, ‘the city is a pattern in time’. It displays a bottom-up collective intelligence that is more sophisticated than the behavior of its parts. The complex aerial choreography that unfolds through the motion of a flock of birds exemplifies the emergence of collective behavior. Underlying the coher- ent elegance and fluidity of the flock is a highly sophisticated form of swarm intelligence premised on the local interaction of individual agents that gives rise to a complex global behavior. The resultant order is not enforced from above, but emerges from the bottom-up interaction of the agents leading to an array of generative architectural design strategies.

29 30 2.7 Swarm Urbanism (Neil Leach)

There are a number of ways of modelling swarm intelligence within a computational framework. Manuel DeLanda outlines a model of agent-based behaviour that could be developed to understand the decision-making processes within an actual city. These agents should be seen as concrete, singular individual agents, and not as abstract agents that embody the collective intelligence of an entire society. DeLanda’s research to date is based on institutional organisations rather than urban forms of the city, and while he envisages the possibility of a model which uses a system of intelligent agents capable of making their own decisions and of influencing others in their decisions in order to generate urban form in some way, he has yet to develop this model. The term ‘swarm urbanism’ has been used fairly extensively within design circles. Often this refers to a form of ‘swarm effect’, where a grid is morphed para- metrically using either digital tools or Frei Otto’s ‘wet grid’ analogue technique. Such techniques, while producing interesting effects, are limited in that

31 they are either topologically fixed (as with a morphodynamic lattice) or base geometrically fixed (as with the wet grid), and cannot make qualitative shifts in form and space outside of these set-ups. The advantage of a genuinely bottom-up emergent system of swarm intelligence where individual agents with embedded intelligence respond to one another is that it offers behavioural translations of topology and geometry that can have radically varied out- puts. One practice that does use swarm intelligence as a fully bottom-up multi-agent design tool is Kokkugia, a network of young Australian architects operating from New York and London. They have deployed this technique at a macro level for a project in the Docklands in Melbourne, an urban redeve opment currently under construction focusing on the extension of the Central Business District into a disused port territory, and have extended it to a micro level with the design of actual buildings, as with their Taipei Performing Arts Centre. With their swarm urbanism projects, the concern of Kokkugia is not to simulate actual populations (of people or institutions) or their occupation of architecture, but to devise processes operating at much greater levels of abstraction that involve seeding design intent into a set of autonomous design agents which are capable of self-organising into emergent urban forms. They are therefore not interested in mapping the motion of swarming agents to generate an urban plan as a single optimal solution, but rather in developing a flexible system embodying a collective self-organising urban intelligence: ‘An application of swarm logic to urbanism enables a shift from notions of the master-plan to that of master-algorithm as an urban design tool. This shift changes the conception of urban design from a sequential set of decisions at reducing scales, to a simultaneous process in which a set of micro or local decisions interact to generate a complex urban system. Rather than designing an urban plan that meets a finite set of criteria, urban imperatives are programmed into a set of agents which are able to self-organise.’ This approach tends to produce a result which – if not reducible to a single steady- state condition – will eventually coalesce into a nearequilibrium, semi-stable state always teetering on the brink of disequilibrium. This allows the system to remain responsive to changing economic, political and social circumstanc- es. Kokkugia therefore sees the urban condition as one of constant flux: ‘Our urban design methodology does not seek to find a single optimum solution but rather a dynamically stable state that feeds off the instabilities of the relations that comprise it.’

32 2.8 Swarm Modeling (Paul Coates , The use of Swarm Intelligence to generate architectural form)

At the paper of swarm modeling, Carranza, Pablo Miranda and Coates, Paul choose swarms as a study case is the fascination of the simplicity of its me- chanics and its complexity as a phenomenon. It can be compared in that sense with other models such as Cellular Automata, for example, with which shares some similarities (they are parallel systems, they interact at a local level, etc). It describes the swarms understanding them as examples of sensori-motor intelligence. It begins addressing some issues already patent when studying simple turtles, and then it looks at two ways of interaction of the swarm and their implications. It studies the interaction with an environment in relation with learning processes and simple perceptions of forms, and then uses the processes developed in this first cases to look at the possibilities of interaction of the swarm with a human, and its similarities with other systems such as Genetic Algorithms or social systems.

33 2.9 Motion at Architecture Design

If motion in arts, montage, program and circulation could be translated into an architecture design method, whether the ground motion like swarm behavior could also generate and architecture design method. Could the movement space of nature could become an architecture space?

34 2.10 Particles at Architecture Design (http://www.heatherwick.com/uk-pavilion/)

Particle elements are used at many architecture examples. Some of these represent the motion pattern of nature movement or events.

At the UK pavilion, Shanghai Expo 2010, predicting that many of the Expo’s pavilions might follow architectural trends in form-making, Heatherwick Studio chosed instead to concentrate on exploring texture. They were thinking of the opening sequence of the 1985 film Witness, in which the camera pans across a field of grass swirled into patterns by the wind. On this windy river- side site, they wanted to make the building’s façade behave like this grass. It also seemed that if you magnified the texture of a building enough, the texture would actually become its form. For the future-gazing expo, seeds seemed an ultimate symbol of potential and promise.

The Seed Cathedral is a box, 15 metres high and 10 metres tall. From every surface protrude silvery hairs, consisting of 60,000 identical rods of clear acrylic, 7.5 metres long, which extend through the walls of the box and lift it into the air. Inside the pavilion, the geometry of the rods forms a space described by a curvaceous undulating surface. There are 250,000 seeds cast into the glassy tips of all the hairs. By day, the pavilion’s interior is lit by the sunlight that comes in along the length of each rod and lights up the seed ends. You can track the daily movement of the sun and pick out the shadows of passing clouds and birds and, when you move around, the light moves with you, glowing most strongly from the hairs that point directly towards you. By night, light sources inside each rod illuminate not only the seed ends inside the structure, but the tips of the hairs outside it, covering the pavilion in tiny points of light that dance and tingle in the breeze.

35 Arne Quinze uses curves, lines, colours and movement in his pieces. his wood- stick structures provide a feeling of movement and fluidity, that combine to create a large frame structure. (http://work-by-djg.blogspot.com/2011/02/artist-research-arne-quinze.html)

“Cities like open-air museums, sounds like realizing my ultimate dream; a confrontation with the public surrounded by art every day. Art has a positive effect on human beings and their personal development; it can extend their horizon and can broaden their view.” ____ Arne Quinze, Sint-Martens-Latem 2011

Quinze is known for his trademark sculptures made out of wooden planks. His installations are built to provoke reaction and to intervene in the daily life of passersby confronted with his sculptures. Quinze sees his installations as places where people meet each other again and start conversations.

36 In 2006, he gained a lot of attention by building Uchronia: A message from the future, a large wide wooden sculpture at the Burning Man festival in Black Rock City,in the Nevada desert, United States. Cityscape (2007) and The Sequence (2008) are two of his giant wooden public art installations in the centre of Brus- sels, Belgium. It was the first time a sculpture gave the impression touching two buildings in the city center while traffic still passes by underneath it. The installation for the Flemish Parliament became an unequivocal actor in the city. In Munich, Germany, he built Traveller (2008) for French luxury fashion and leather goods brand Louis Vuitton. Other public art installations by Arne Quinze have recently been revealed in the centre of Paris, France (Rebirth, 2008),Beirut, Lebanon (The Visitor, 2009)and Louisville, Kentucky (Big Four Bridge, ongoing).

Arne Quinze: “With these sculptures I’m looking for a confrontation with the public, I hope they start asking questions about what their function on this planet is. What happens when putting all of the sudden an alien element in the city, our habitual urban environment? How do we react to unusual objects if we are confronted with them in our daily lives? Who or what remains the stranger, the person confronted with it or the object itself?”

37 GC Prostho Museum Research Center, by Kengo Kuma, is architecture that originates from the system of Cidori, an old Japanese toy. Cidori is an assem- bly of wood sticks with joints having unique shape, which can be extended merely by twisting the sticks, without any nails or metal fittings. The tradition of this toy has been passed on in Hida Takayama, a small town in a mountain, where many skilled craftsmen still exist.

The cement structure of the GC Prostho Museum Research Center has a rect- angular floor plan on 3 levels surrounded by a parametric decorative system formed of cypress wood elements generating regular prismatic combinations created with interlocking joints.

This architecture shows the possibility of creating a universe by combining small units like toys with your own hands. We worked on the project in the hope that the era of machine-made architectures would be over, and human beings would build them again by themselves. (http://www.archdaily.com/199442/gc-prostho-museum-research-center-kengo-kuma-asso- ciates/)

38 39 If the swarm behavior can be visually seen as some layers of architecture skin, then patterns and layering might be a way of application in architecture design for swarm simulation. There are something new and unseen in preexist- ing research on Japan about patterns and layering by Liotta and Matteo.

Salvator-John and Matteo have attempted to create a link between patterns and layering. These two previously detached notions can now be integrated into one methodology mediated by structural concepts that are the key to this link. Structural analysis of the twentieth century struggled to advance beyond the column and beam structural frame. Analysis today allows us to conceive stable structures through the accumulation of delicate members, which have the capacity to produce a variety of patterns while fulfilling their structural responsibilities. (Salvator-John A. LIOTT A and Matteo BELFIORE, Patterns and layering)

40 Chapter3: Swarm Tectonics

41 3.1 Swarm Testing

By using meatball and lines, swarm could be used as a tool of form finding.

42 43 3.2 From Simulation to Application

Birds flocking and fish schooling were simulated by digital animation tool and then replicated by different kind of particle elements. It was visually tested for what kind of space could be resulted.

44 45 Rather than static space constructed by huge concrete, glass, cubic frames, the space formed by swarm might be constructed with more natural, various sized materials and there are more potential for the skin which might be in between solid, transparent, or translucent.

46 47 48 49 50 3.3 Swarm structure

If swarm was seen as motion of particles, the connection by the particles or the connection between different paths could construct as a structure-like system.

51 52 53 54 55 56 Tracking and connection are the two points to construct this kind of structure system. The space could be treated as a mobile zeppelin above the city or a pavilion under the water.

57 58 59 3.4 Skin attachment

Material characteristics of the skin decides the form. There are two types of skin. One is frame with membrane, the other is panels connection.

60 61 62 3.5 Swarm Modular

A modular space could be seen as a cube controlled by 8 swarm points. Form was changed based on the movement of the points.

63 Simple xyz movement, rotation and extension, could generate complated morph system.

64 Swarm simulation and robot control could translate data into architecture form.

65 Connection enable the simple modular to construct various geometry.

66 67 3.6 Swarm Joint

Rotation, extension and connection are three main typical mechanism design.

Rotation

68 Connection Extendsion

69 Three type of joint could construct a swarm modular.

70 71 Flexible material is casted into an organic shape outside of the mechanism system. They work as a relationship like bone and muscle.

72 73 3.7 Swarm Drawing

Swarm modular could generate a basic architecture frame. Particles around will swarm based on simple principles to generate a form.

74 If the path of these particles could be seen as a sketch drawing for architecture sketch, fibrous system could be used to simplify the complex lines into a skin frame drawing.

75 Attractor

Attracting region

Attractor path Predator Predator boundary Bounding edge

Test No.01 Test No.02 Test No.03 Test No.04 Test No.05 Test No.06 Test No.07 Test No.08 Test No.09

76 Seek force : 172 Seek force : 190 Seek force : 90 Predator force : 251 Predator force : 251 Predator force : 251 Allignment : 251 Allignment :251 Allignment : 251 Cohension : 1000 Cohension : 1000 Cohension : 1000 No.01 Seperation : 148 No.02 Seperation : 148 No.03 Seperation : 148

Seek force : 193 Seek force : 193 Seek force : 193 Predator force : 196 Predator force : 187 Predator force : 251 Allignment : 528 Allignment : 528 Allignment : 528 Cohension : 737 Cohension : 737 Cohension : 737 Seperation : 444 Seperation : 444 Seperation : 444 No.04 No.05 No.06

Seek force : 163 Seek force : 163 Seek force : 163 Predator force : 200 Predator force : 200 Predator force : 200 Allignment : 323 Allignment : 300 Allignment : 251 Cohension : 456 Cohension : 456 Cohension : 456 Seperation : 130 Seperation : 130 Seperation : 130 Swarm ParticleDrawings No.07 No.08 No.09

77 No.01 No.02 No.03

No.04 No.05 No.06

Reconstruct No.07 No.08 No.09

Simplify

78 No.01 No.02 No.03

No.04 No.05 No.06

3D modeling No.07 No.08 No.09

Modeling

79 Steel Sketch Model

80 A D C B

LEVE 3

LEVE 2

LEVE 1

1:100

A B C 1m 3m 6m 9m

Elevation

81 3D Priint Frame Model

82 3D Priint Frame Model

83 3D Priint Skin Model

84 3D Priint Skin Model

85 Steel Sketch Model

86 C D A

LEVE 6

LEVE 5

LEVE 4

LEVE 3

A LEVE 2

LEVE 1

B LEVE 0

1:50 C D 1m 3m 6m 9m

Elevation Drawing

87 3D Priint Model

88 3D Priint Model

89 3.8 Data Swarm

All particles active in groups and interactive with each other. For fish and birds, the predator and attractor acts like a main force to effect the space created by swarm. At the digital simulation, all points were set up with swarm properties, effected by different kind of forces.

Top Urban area for startups Startup Per Year, 1-5 (1995-2014) Startup Per Year, 5-15 (1995-2014) Startup Per Year, 15-250 (1995-2014) Investment Flow Lines Investee cities

Main Startup cities New Startups Per Year Investment Flow from San Francisco Investment Flow from New York These forces could be wind, temperature, light, activities of human, program of functions, or circulation. Therefore, we can translate the mapping of forces into a kind of constrain for architectural swarm.

Investment Flow from San Francisco Investment Flow from New York The migration flows of startup investment acts like animals swarming in between cities. Events for startups happen around bay area with the increasing of new companies. Through the simulation of these events, an urban swarm footprint was discovered.

Redevelopment Agency District Office Downtown Office

Downtown Commercial Production, Distribution & Repair District

Neighborhood Comercial Residential-Comercial Combined Residential, Mixed (Houses & Appartments) Residential, Houses

Residential, Houses Open Space

Main Roads and Highway Bicycle Network Off Street Parking

Startupw with fundings

2014-2015 Startup Event Locations

Restaurants

Esri, HERE, DeLorme, MapmyIndia, © OpenStreetMap contributors, and the GIS user community An architecture prototype based on swarm behavior and effected by the architectural forces will generate a new kind of incubator space for the startup companies migrating above the cities.

Building height: 20-32 feet

Building height: 33-50 feet

Building height: 51-70 feet Building height: 71-96 feet

Building height: 97-125 feet Building height: 126-220 feet Building height: 221-360 feet

Building height: 361-1000 feet Open Space

Main Roads and Highway Bicycle Network Off Street Parking

Startupw with fundings

2014-2015 Startup Event Locations Restaurants

Esri, HERE, DeLorme, MapmyIndia, © OpenStreetMap contributors, and the GIS user community Total funding for starups (2004-2014) Total funding for starups (2004-2014)

Investment flow simulation Investment flow simulation

94 Time frame 0 Time frame 3

Time frame 1 Time frame 4

Time frame 2 Time frame 5

95 96 3.9 Conclusion

Swarm could be generated into an architectural design process. Data constrain and the artificial selection are two of the key points. Data constrain rep- resents a digital environment constructed by the data information from natural and social environment. Artificial selection is based on architects’ decision. Different from mass production, there aren’t two identical form generated by swarm, but the initial swarm principles and process are simple and similar. Four main technology for swarm tectonics are mechanical robot control, 3d printing, data collection and swarm simulation.

Swarm tectonics will increase the productivity and identity of architecture. Data information will generate a digital environment as constrain for swarm architectural behavior. Rather than designing a form, architects design a process to generate the form by swarm tectonics.

97 Bibliography:

1. Steven Johnson, Emergence

2. Daniel Schodek, Dynamic digital representations in architecture visions in motion

3. Smart Swarms – How Understanding Flocks, Schools and Colonies Can Make Us Better at Communicating, Decision Making and Getting Things Done

4. Craig W. Reynolds, Flocks, Herds, and Schools: a Distributed Behavioral Model

5. Carranza, Pablo Miranda and Coates, Paul, Swarm modelling

6. Lebbeus Woods, Radical Reconstruction

7. Rem Koolhaas, Delirious New York: A Retroactive Manifesto for Manhattan

8. Sou Fujimoto , Primitive Future

9. Theodore Spyropoulos, John Frazer , Patrik Schumacher, Adaptive Ecologies