This summer we are undertaking a fresh investigation into the now classic subject of fabrication methods for developing pattern/gradient on a double-curved surface.  While this project builds directly on methods developed for Robofold instead of looking at panelization, we are focusing developing the inherent properties of the form itself.

We are currently based in Lexington, Kentucky and this gives us an infinite supply of complex forms to test on, but they happen to be covered in grass.  This project thus looks at developing the lawnmower as a method for fabricating relief patterns across these surfaces.  Mowers have been used for decades as a means of creating 2D patterns for simple front lawns and sports fields, and have been the focus of other architectural enthusiasts such as Archigram.  This research examines the results of adding CNC control the mower to transform it into a 3D printing mechanism.  Since it is still a push-mower it also examines its interface to the operator to achieve the correct speed and angle for each print line.

The initial prototype is constructed as a rolling frame which holds the mowing mechanism.  The mowing blade is taken from a standard reel mower and is controlled by 2 linear actuators using an Arduino.  These actuators provide dynamic control of both the height and angle of the blade with a range of 6”.

Project Team: Anton Bakerjian,  Matt Nett

We have recently installed our Mits Eleven Lab CNC.  This means we now have the capability to produce custom PCBs in house.  First output soon.

Last Friday the College of Design opened its end of year show at LOT. The show featured work from the graduate studio I co-taught with Erick Carcamo of X|Atelier.

Over a 2 week period the World Equestrian Games will increase the population of Lexington by approximately 400% and radically alter the city’s notion of infrastructure, commerce, and circulation. In response to this scenario, the studio undertook an investigation into the relationship of excessive digital form making and manipulation via interactive devices to develop temporary structures and programs throughout the city. The methods examine how sensing and tracking technologies can augment the design process and extend the potential for architects to deal with extreme scenarios.

Students:

Anton Bakerjian
John Haas
Christopher Harris
William Hart
Jonathan Lee
Matthew Long
John Mason
Heather Micallef
Bobby Morris
Matthew Nett
Andrew Owens
Myra Shuffett

AltnResearch recently completed a collaboration with Robofold to develop a robotic folding system for Studio Joris Laarman. The system of 4 robots simulates production of Laarman’s Asimov chair by folding it from a flat sheet of metal. The prototype is currently on show at the Friedman Benda Gallery in New York City. - more info soon

Applicants should preferably be familiar with 3D modelling/animation software (particularly Maya) and have a knowledge of both software and hardware programming.  Interested candidates should send a CV/brief portfolio (No files larger than 5mb) to:

internships{AT}altnresearch.com

About SimAud:

Symposium Chair

• Azam Khan (Autodesk Research)

Program Committee

• Robert Aish (Autodesk)
• Ramtin Attar (Autodesk Research)
• Ellen Yi-Luen Do (College of Architecture & College of Computing, Georgia Tech)
• Mark D Gross (CoDe Lab, Carnegie Mellon University)
• Kasper Hornbæk (Department of Computer Science, University of Copenhagen)
• Michael Jemtrud (School of Architecture, McGill University)
• Judit Kimpian (Head of Sustainability and Advanced Modelling, AEDAS)
• Branko Kolarevic (Faculty of Environmental Design, University of Calgary)
• Barry Leonard (Air Staff, National Defence Headquarters)
• Greg Madey (Research Professor, University of Notre Dame)
• Rivka Oxman (Vice Dean, Faculty of Architecture and Town Planning, Technion ITT)
• Gabriel Wainer (ARS Lab, Carleton University)

Unlike other research conferences and symposia that merely provide a venue for publishing research on a particular topic, SimAUD will work to build a closer community for cross-pollinating domain knowledge, collaborative work and co-authoring of research that envisions the future of the built environment. Important to this work is understanding and modeling our current built and natural environments. From there, we can validate our simulations against existing physical systems. Once validated, we can have confidence that simulating new designs or design alternatives, will accurately compute outcomes. Therefore we encourage the submission of case studies documenting lessons learned so that SimAUD can also be a resource of best practices. Papers may also propose extensions to modeling and simulation standards, such as IFC, gbXML, LandXML, and CityGML, to advance progress, or identify and propose new standards that are needed. We encourage members of the community to submit models of their own buildings, campuses, or cities, and to participate actively as leaders in local initiatives and solutions (which make great case study submissions too).

Simulation is a “game-changer” in architecture. We now need to minimize the impact of human activity on the environment, especially in the face of an increasing population. Within this scope, the built environment is the most significant contributor to the problems of climate change. The built environment is a complex system, and planning for change requires high-quality modeling and simulation. Single buildings are key to making change, but urban design must also be considered. By submitting your original research to SimAUD, you contribute directly to the global effort to untangle the complexities of urban development and mitigate the negative effects of human activity on the planet.

Overview

The Xiamen Energy Masterplan Model was developed with CHORA as part of their scheme for the Taiwan Strait Incubator project.  This portion of the scheme looks to establish Xiamen as the first “Green” or Energy Efficient City in China.  To achieve this the city commissioned an overall “Energy Masterplan” from CHORA which addresses this goal at a large scale through large infrastructure and building projects, as well as local interventions by residents.  The model looked to address the need to visualize individual projects in the context of the overall scheme to both educate the public and act as a planning tool for developers, designers, and city officials.  This is achieved by combining unique animations in conjunction with detailed screen-based information to both understand the current proposal and serve as a mechanism for developing it further.

Overall View of the model

Visualization

The concept of an Energy Masterplan is fairly new so one of the basic tasks was to develop a method that would allow the different schemes to be visualized spatially and with detailed information.  To do this, the city was modeled at 1:10,000 and printed as 122 rapid prototype SLS prints.  The tiles are stitched together to form a 4MX4M model of the city.  The energy information is then overlaid onto the model via 620 leds which backlight the SLS tiles, using them as a diffuser.

Typologies / Color / Interaction

The Masterplan is divided into 4 typologies and each of these are represented by a different color LED in the model:  Energy Networks-Blue, Energy Efficiency-Yellow,  Renewable Energy-Red, and Pilot Projects-Green.  Each of these categories were divided into 5 areas:

Energy Network: Clean Transportation, District Cooling, GreenTechnology, Intelligent Energy Network, Local Smart Grid

Energy Efficiency:  Efficient Coal Fire Power, Grey Water Recycling, Home Improvement Kits, Low Impact Construction, Passive Building

Renewable Energy:  Algae Biomass, Energy Façade, Energy Island, Energy Tower, Solar Power Plant

Pilot Projects:  Commercial Building, Energy Museum, Energy Road, Industrial Headquarters, Waste Treatment

In this same way the interface to the 4 types of projects was achieved via a terminal located at each corner of the model.  Each terminal has 5 simple button inputs which correspond to the 5 subtypes.  If no buttons from the terminal have been pushed recently the types revert to their animations of the “musical score”  This musical score acts as a passive, narrative interface to the various subtypes.  Once any button on a terminal is pressed, the pressed button triggers it “active” animation sequence and the other subtypes trigger their “background” animation, which makes it easier to distinguish the subtype which has been pressed.  Each subtype also has a unique pulse animation which serves as the mechanism to indentify it within the overall typology.  Adjacent  to each of the terminals is a screen which displays detailed information regarding the project that is active on the terminal.  This sets up a complimentary relationship between the spatial information overlaid onto the model and the detailed data displayed on the screen.

Construction Methodology

To allow the model to be easily transported it is divided into 16 1mX1m units.  The challenge that this presents is that most of the circuits of a given typology span between 2 and 14 of these units.  To account for this, each unit has a local connection board which houses all of the individual circuits on the block.  To connect the circuits together a series of jump connectors were developed which span between the blocks and also allow for people to pass between the blocks for the installation.  To create the connections the blocks are assembled into 4 columns of 4 block each with a walkspace between them.  Then using the labels and plans an individual simply follows the circuit and plugs into the appropriate connector at each block.  Once all the circuits are connected and tested the blocks are brought together into a continuous surface.

Light/Screen Controllers

Each of the 4 terminals is attached to an Arduino Mega which is wired to a custom circuit which handles the Input, Lighting Control, and Screen Information.  In its final form, the controller is wired to simple barrel connectors to allow simple assembly connection of the lighting circuits and button inputs.  It is also equipped with basic cooling to keep the system stable when running for long periods of time.  Each controller is also linked to a PC which is running a Flash page.  As the Arduino adjusts its lighting animations based on the inputs, it also sends information to flash about which page to display.

The main innovation in the software is the ability to carefully craft the animation in the same way that you would edit a film.  Each circuit has 3 distinct “animations”: Music Score, Background, and Active.  Each of these sequences are defined through pairs of values which define a desired brightness and the amount of time taken to achieve the brightness.  Each animation can contain as many defining steps as required (a minimum of 2) to allow the animation to distinguish it in the model.  Once an animation reaches the end of the sequence, it loops back to the beginning value.

For example:

If a background animation is defined with the pair of arrays

backG_brightness[]={0,255};

backG_times[]          ={100,1000};

The circuit would start at 0 (completely off) and fade up to 255 (maximum brightness) in 100 milliseconds, then fade back to 0 in 1 second.

But if you wanted to control the animation more closely, you could add more brightness/time pairs.

background_brightness[]={0,255,50,100,0,30};

backgroung_times[]          ={100,1000,20,50,100,700};

The software is designed to be flexible and allow any number of brightness/time pairs per animation.  In this way each circuit is embedded with 3 distinct behaviors.  These behaviors can be easily altered allowing the model to take on different personalities depending on it audience/users at the time.

1 of 4 controllers, powered by an Arduino

Initial Tests

Future Development

The main developments with the model lie in the enhancement of the “local interactivity” and the “global connectivity” of the model.  Though intentionally simple at the moment – the button press input could be extended to allow users to interact with a capacitive surface which could also act as a display mechanism as the model changes modes.  The other major development area will focus on the “global connectivity” of the model.  At the moment, the model relies on predefined animations and presentations to visualize the various components of the masterplan.  Future developments will link the model to live data over the internet such as current energy prices or local pollution conditions, link it with a corresponding project within the model and alter the display of the lighting patterns and screen information.  In this way the model could be adapted to each of the cities venues it visits by comparing local conditions with that of Xiamen, and would become an active display and planning tool which learned more information as it travelled.

The new and official home for all research into curved folding as a method formal research and new manufacturing techniques.  It is being run by Greg Epps of Robofold.  I’ll be contributing to the site via my ongoing research into these methods