IEEE Games Entertainment and Media 2019 - Published paper
Sound Space utilizes mixed reality technology to augment the process of designing spaces for acoustic efficiency by simulating sound generation visually.
In this work in progress I am focusing on using motors to expand and contract a matrix of flowers using capacitive touch sensors + arduinos. In an effort to communicate with pollinators, some species of flowers will close after a pollinator has made contact with it. This project aims to educate viewers about that relationship within an NYC butterfly/insect exhibit.
(Work done in conjunction with Terreform ONE research fellowship)
The goal of the Breathing Plaza is to create a space that reacts in real time to use. If a person enters the plaza, the structure will slowly envelop them, creating a bubble-like retreat from the outside. If a group of people enter, a larger bubble will form - creating an intimate meeting environment. The adaptations of the plaza are telegraphed visually to passersby and users within, allowing them to either join or avoid.
I conducted multiple experiments on both the actuators themselves and the connections of these actuators into a larger system. Each actuator is composed using similar concepts and components - an inner balloon is encased within a diamond braided mesh fabric - which when inflated restricts the normal expansion of the balloon instead into linear contraction.
Later experiments build off this concept - switching out the balloon for a silicone bladder and the mesh for a ribbon (braided with the same diamond pattern to create the necessary contraction). An inextensible material - in this case a simple strip of paper - was also introduced in an effort to change the direction of the contraction.
Two systems came of these experiments - the first using the pneumatic muscles (experiment 1) to pull on an elastic fabric to change the shape of its surface (experiment 7), and the second using the chained combo actuators (experiment 6) to create a surface made completely of the actuators themselves (experiment 8).
Companion Cubes is an interactive installation that utilizes lights, arduinos, and photo-sensors to engage users in collaborative play. With the shape and scale of a children’s building block, but the aesthetic of a contemporary lamp, Companion Cubes encourages both children and adults to create unique forms and structures with the toy-like objects.
Users interact with the installation by creating and breaking chain reactions of light with the individual lamps. A light chain is activated by using an external light source – like a flashlight or a phone – which then triggers any boxes within the predetermined range of sensitivity. Because of their rectilinear shape, lamps can be placed in rows, stacked, and placed at random. In presentations of this project, users have immediately experimented with how the chain reaction works and each lamp’s range of sensitivity, and then proceed to collaborate with the rest of the group to create interesting forms and reactions. A phone light can also be shone across multiple lights at once, allowing users to interfere with and reset chain reactions. Breaking the chains can also be fun for the group, with users working together to frantically break the chain reaction while the lamps work against them.
Hardware: For the hardware, an arduino/microcontroller is attached to a soldered perf board that contains 3 LEDs as its output and 2 photoresistors as its input. A 9V battery is also connected to the arduino as its power source.
Software: The code is uploaded from Arduino 1.8.5 onto the microcontroller which tells the arduino to receive the photoresistors as 2 separate inputs, to trigger all 3 LEDs with a fade-in and fade-out effect when either photoresistor receives an input above a certain threshold – which is calibrated to the ambient light of the current room of the project – and to trigger the lights at a randomized delay. The randomized delay occurs within a range of .5 to 1 seconds and prevents the lamps from reacting to themselves, which would cause infinite loops, instead of a chain reaction. The delay is randomized instead of constant to create a twinkling effect within the chain reaction.
The Magic Carpet is a project created in the Institute for Advanced Architecutre of Catalonia, during the two-term seminar Mushrooms: Data Informed Structures.
The Magic Carpet is composed of wooden cross sections that create folded surfaces, which can be used as seating, lounging or working spaces. The chief issue we faced was making sure that the origami-like 3D surface would not “unfold”, instead ensuring that it self-locked in place. Additionally we focused on ergonomics and ensuring that an angle between 90 and 160 degrees was maintained to increase the user’s comfort, whilst keeping in mind that the steeper the slope, the less it unfolds, thus the sturdier the structure. The “magic” of the magic carpet is that it geometrically/ morphologically spans along one axis, while it structurally spans along the other, causing an inversion of expectations.
Initially, the structure was going to be installed in the IAAC facilities as a new lounge space between two existing bridges. Later, it was decided to turn the Magic Carpet into a mobile structure, let it fly and give the public space a new identity, allowing the users of the city of Barcelona to interact with it and seize their own space in the diverse forms it offers.
“We believe that, if human beings are part of an ecology, then the objects humans make should also be part of it. Among humans and insects alike, inhabitable spaces are the result of a deliberate organization of material, energy, information and a continuous interaction with the environment, whose goal is to help develop tight-knit communities.”
The purpose of this project is to design a photobioreactor in an urban environment for the production and consumption of future food - specifically the micro algae spirulina.
The shape of the brain coral lends itself readily to the form of a tubular photobioreactor, with winding hills and valleys representing the tubes themselves. Brain Corals exist symbiotically with algae; the algae depends on the stony coral for structure while the coral depends on the algae for food.
A brain coral pattern can be generated using a couple of different algorithms, particularly through the concepts of reaction-diffusion and differential growth. Reaction-diffusion is pixel or grid based simulation that forms all at once, instead of gradually from an initial position – which does not lend itself to the functionality of a tubular photobioreactor. Using the concept of linear differential growth, we were able to design a tube system that was at once continuous and variegated - in a controlled way.
In this case, differential growth is generated using a mass-spring system - a physics simulation that treats lines and surfaces as springs in tension. This algorithm, based on the work of Daniel Piker, works through line expansion growth. An initial line is segmented into individual springs and grows through lengthening each spring individually with collision detection. Spheres drawn on each point of division provide the collision detection, and as a result a hexagonal organization will naturally occur (circles tend to organize into a hexagonal grid).
The photobioreactor was fabricated with hand-blown glass by a local glass artist. The glass sculpture is then backed with an Arduino controlled LED panel that pulses with the same timing as a heartbeat. The time increments between the LED flashes is decreased when approached - through the use of motion sensors. The differential widths of the glass channels were designed to optimize the flow of spirulina, with smaller widths increasing flow speed upward against gravity and larger widths only in areas where gravity assists the water flow. For its exhibition at IAAC, the photobioreactor was mounted on 3 wooden panels that were CNC milled to show the continuation of the pattern as it would be in an ideal scenario.
Examples of parametric design work, done in Grasshopper
“The idea or concept is the most important aspect of the work; the planning and decisions are done beforehand and the execution is a perfunctory affair. The idea becomes the machine that makes the art.”- Sol Lewitt
Through the use of a simple, pure language of color and form, a series of rule sets were designed that replicated desired conditions. These rules were then rendered into visual, and then physical, representations. The rules evolved, growing more complex with every variation, affecting every aspect of the design, even the door hinges. The initial set of rules below generated the pattern in this spread:
1. Each black square must be connected to at least one black square and one white square.
2. There can be no more and no less than 30 black squares connected to each other at a side, unless there is no room for all 32 squares. This is called a cluster.
3. Each cluster must be separated from each other cluster by at least one white square.
Here I generated Sol Lewitt’s Wall Art as a script in grasshopper to demonstrate its differences to a biomimetic script. From Wikipedia: A cellular automaton consists of a regular grid of cells, each in one of a finite number of states, such as on and off. The grid can be in any finite number of dimensions. For each cell, a set of cells called its neighborhood is defined relative to the specified cell. An initial state is selected by assigning a state for each cell. A new generation is created, according to some fixed rule that determines the new state of each cell in terms of the current state of the cell and the states of the cells in its neighborhood. Some biological processes occur—or can be simulated—by cellular automata, like the patterns of some seashells.
Realizing that the original rule set shared a surprising number of traits to the mathematical theory of “cellular automata”, I translated the written procedure into the biomimetic code. Building off of a cellular automata code, I generated a grasshopper script that looped through the predefined rules. Black squares became activators (on), white squares became inhibitors (off), and clusters became neighborhoods. The more times the loop repeated, the more defined the neighborhoods became. My original idea resembled the output generated by the second run most closely, while the tenth run generated unexpected oblong shapes. My manual reproduction of the rules was ultimately hindered by my inability to draw the whole at once; I had to start at one point and finish at another, while the script could run through each rule simultaneously.