Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.



An algorithmic simple-ruled complexity for generating a flexible urban design tool

serdar aydin

on 5 September 2012

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of quasiGRAMMARS

Urban Grammars Jose Duarte Programming

ARCH 419 MA thesis quasiGRAMMARS dissertation Serdar AYDIN An algorithmic simple-ruled complexity for generating a flexible urban design What's wrong with urban !.. Computation for Liveable Urban Shape Grammars Genetic Algorithms 3. specific solution individual elements to depend on each other If Order != Disorder

then Order != Randomness which is a case of disorder Complexity Planning & philosophy Fast Growing vs.
Complexity 1. problem in the broad context 1. problem in the broad context Top-down approaches Urban pattern Urbanisation along the history introduction Paradigm Shift
needed 2. solution in the broad context Timeline for computation 1. problem in the broad context Evolutionary algorithms Computation & randomness Complexity & randomness Criteria for success Richness of randomness!.. Large-scale
urban projects 3. specific problem Different form of Neoliberal
development: China Building future's of cities
by TOKI: Turkey (Re)developing an urban Mozart Stacking blocks in the name of development!.. Genetic code-scripting Bottom-up approaches and Girih Fractal strategy GAs GPs Genetic Algorithms Genetic Programming High-rise shortcuts why is vernacular complexity so expensive?.. How can we simply do it again?.. The answer is... Optimization of Shape Grammars by Genetic Algorithm We will use Processing... 1st step Finished 2nd step Fitness Last step Start why how to design a contextual urban pattern for to be optimised by EA techniques generate random population a search technique for optimising to find possible solutions Criteria for success a kind of Evolutionary Algorithms So.. How does it work! Is that all? Genetic Programming Is there a better way?.. What should not be expected? What should be expected? by Frazer, J. to replicate accurately
variety and mutation by crossover
capability for reproduction
massive overproduction
selective competition Basics of Algorithms Basic linguistic elements constants
libraries Basic operations arithmetical
classifications class Population {

} float mutationRate; // Mutation rate DNA[] population; // Array to hold the current population int MAX; // Population maximum float [] fitness; // Separate array to hold corresponding fitness value ArrayList darwin; // ArrayList which we will use for our "mating pool" String optimum; // Target int generations; // Number of generations boolean finished; // finished evolving? Evaluate the fitness of each genotype void calcFitness() {
for (int i = 0; i < population.length; i++) {
fitness[i] = population[i].fitness(phrase);
} Generate a mating pool New population Genotypes Selection
of two parents according to their fitness Crossover 3rd step Mutation 4th step Accepting Replace Use the new population for a further run Test If satisfied, stop and return the best solution Go to Step 2 DNA void naturalSelection() {
float totalFitness = getTotalFitness();
for (int i = 0; i < population.length; i++) {
float fitnessNormal = fitness[i] / totalFitness;
int n = (int) (fitnessNormal * 10000.0f);
for (int j = 0; j < n; j++) {
} void crossover() {
for (int i = 0; i < population.length; i++) {
int m = int(random(darwin.size()));
int d = int(random(darwin.size()));
DNA mom = (DNA) darwin.get(m);
DNA dad = (DNA) darwin.get(d);
DNA child = mom.mate(dad);
} child.mutate(mutationRate);
population[i] = child; for (int i = 0; i < population.length; i++) {
population[i] = new DNA(phrase.length());
} Placement of the solution in a new population Loop to have a manageable complexity to be pointed out from the existing pattern to be complied with Flexible Urban Design Tool But Order Randomness Algebra and Arithmetic Instead of euclidean geometry Because Randomness == max Complexity

Randomness != Regularity && Simplicity & Randomness INDIVIDUALITY


IDENTITY COMPLEXITY every change to require the whole newly constructed flexible dynamic systems in balance with programmed computers more fluid, more ductile architecture Scripting & Codewriting Languages:

C, C#, C++
Processing (Java)
RhinoScript (VB, Grasshopper)
AutoLISP (AutoCAD) Simple yet powerful
Can be integrated into Grasshopper No, thanks!.. generated in
AutoLISP (AutoCAD) Some algorithms to address design issues:
Voronoi Tessellation
Stochastic Search
Cellular Automata
Evolutionary Algorithms GAs GPs The process derives from GAs'. Then what is the difference? 1. Tree-like genotype, no binary codes
2. Terminals and functions
3. Pressure on DNA An Example for a Combination of GAs and Shape Grammars Manual Design Techniques Complexity Computational Optimisation START CROSSOVER WAY GENOTYPE Offspring Offspring Member Member Member Member Member Member Member branches are functions fruits are terminals GP machine GA's crossover http://www.geneticprogramming.us/Genetic_Operations.html easier and faster;

creation of large urban scenarios with the help of the method by which genetic algorithms and shape grammars integrated

generation and visualisation of large-scale projects

integration of non-regular geometry

manageable complexity for complicated urban design problems

reusability of design patterns for sustainable urban design


may come to an abrupt end inadvertently

may be misunderstood!! Processing codes exportable to Rhino/Grasshopper Make a tool thank you... "a goodly word like a goodly tree" quasiGRAMMARS Further reading: Aranda, Benjamin and Chris Lasch. 2008. "Out of Order." In From Control to Design. Barcelona: Actar Duarte, Jose P., Joao M. Rocha and Gonçalo Ducla Soarez. "Unveiling the structure of the Marrakech Medina: A shape grammar and an interpreter for generating urban form." Terzidis, Kostas. 2009. "Algorithmic Architecture." Oxford: Architectural Press. Terzidis, Kostas. 2009. "Algorithms for Visual Design Using the Processing Language." Indianapolis: Wiley Publishing. Coates, Paul. 2010. Programming.Architecture. New York: Routledge Conclusion 1970 by Holland 1991 by Koza 1972 by Stiny & Gibs
Full transcript