This section will serve as your report to your program officer of your project's activities and findings. Please describe what you have done and what you have learned, broken down into four categories:
1. Describe the major research and education activities of the project.
CBA's research program is studying the foundations of how a logical description of a system can be embodied in a physical representation, and the converse. These are very old questions, dating back to the emergence of modern engineering practice and beyond that to the origins of the natural sciences; what is new in both directions are the remarkable consequences of recognizing and applying the computational capabilities of physical mechanisms:
- Creating physical forms
with desired functions
The possibility of a computationally-universal assembler detecting and correcting its own errors promises to do for fabrication what a digital representation has done for communications and computation, introducing a threshold to allow perfect macroscopic structures to be assembled from imperfect microscopic components.
- Abstracting functional
descriptions from physical forms
A microscopic specification of the information processing in a thermodynamically-complex system promises to lead to not just a description of its macroscopic behavior but also the means to reliably direct it, without requiring a detailed specification of its internal configuration.
CBA is training students at MIT to work in these emerging areas through new interdisciplinary classes that have been developed with a strong project focus, and, beyond MIT, CBA's field "fab labs" are bringing emerging capabilities for personal fabrication out of the laboratory and into under-served parts of the world in order to help address what can be thought of as instrumentation and fabrication divides.
2. Describe the major findings resulting from these activities.
The original CBA proposal outlined research in three layers: Foundations of physical mechanisms for manipulating information, their integration in computational Substrates, and their organization in turn in large-scale Systems. The results from the preceding year represent the maturation of these areas into three kinds of emerging experimental practice that could be called "quantum-mechanical" engineering, "computational-mechanical" engineering, and "statistical-mechanical" engineering:
- "Quantum-Mechanical"
Engineering
CBA experiments in quantum information processing have sought to program naturally-occurring mechanisms that already preserve quantum coherence while providing controllable interactions. In the last year there have been results in the representation of information and algorithms, in their scaling, and in appropriate architectures:
- Representation
- Adiabatic algorithms, based on continuously deforming a
Hamiltonian from one with a trivial ground state to one that encodes a
desired computation, are theoretically equivalent to the standard
quantum circuit model, but are very different experimentally. The first experimental implementation
of an adiabatic algorithm was accomplished, solving a max-cut
optimization problem [Experimental
Implementation of an Adiabatic Quantum Optimization Algorithm, M.
Steffen, W. van Dam, T. Hogg, G. Breyta, and I. Chuang, Phys. Rev. Lett. (90) #067903
(2003)]. It was observed that the correct answer is still found even if
the Hamiltonian is modified too quickly, shown here in state tomography
on the density matrix:
(where M is the number of approximating steps used). This approach offers a new way to generate approximate quantum algorithms; one of the most important classical computing results has been the power of approximate algorithms that work well on typical cases but not on worst cases. A goal for the coming year will be to study the possibility of using adiabatic "fast" passage for speeding up approximate quantum algorithms.
- Spin-based quantum computations to date have used
spin-1/2 qubits, however quadrupolar moments offer the possibility of
representing more information per particle. This was demonstrated in the first spin-3/2 quantum computation,
using Cs133 in cesium pentadecafluorooctanoate. This was
accomplished by mapping the spin-1/2 circuit into a spin-3/2
representation:
using a new technique to correct the phase errors from Bloch-Siegert shifts [M. Steffen, K.V.R.M. Murali, P. Judeinstein, I. Chuang, submitted].
- Along with improving computational performance,
physical mechanisms offer new ways to protect information security.
Photon scattering from inhomogeneous materials was used to introduce a physical one-way function
by showing that in the mesoscopic limit it is equivalent to a one-way
hash of the scattering structure [Physical One-Way Functions, R.
Pappu, B. Recht, J. Taylor, N. Gershenfeld, Science (297) pp. 2026-2030 (2002)].
- Scaling
Spin-based molecular computing in thermal liquids has led other approaches to quantum computing in problem size and complexity, but it will not scale beyond tens of qubits because of the exponential fall-off of the partition function normalization. Two complementary approaches are being investigated to address this limit:
- CBA's nanolithography tools are being used to write electronic interfaces for single molecules; a goal for the coming year will be to electronically observe the spin state of a single molecule.
- Ultra-fast optical pulse shaping is being used for
pumping molecular electronic states. An initial result has been the first demonstration of optical quantum
control of spin dynamics, selectively modifying the
singlet-triplet branching ratios with chirps found by machine learning
to suppress and enhance it (left and right Wigner plots):
[J. Taylor, P. Bucksbaum, N. Gershenfeld, to be submitted].
- Architecture
In CBA's studies of quantum information architectures, entanglement emerged as a key resource in two new areas:
- Entanglement was
proved to provide a sqrt(N) improvement in the power-limited
communication capacity of a coupled channel [Power of Entanglement in Quantum
Communication, S. Lloyd, Phys. Rev. Lett. (90) #167902 (2003)].
- Nearest-neighbor swap operations have been frequently
assumed in quantum computing proposals, but the overhead of doing this
reliably was shown to not scale; instead, teleportation was proved to scale
interconnect in large quantum computers [A Practical Architecture for Reliable
Quantum Computers, M. Oskin, F.T. Chong, I.L. Chuang IL, Computer
(35) 79-+ (2002)].
- "Computational-Mechanical"
Engineering
The second broad layer of activity in CBA has been an experimental and theoretical study of the role of computation within physical fabrication, seeking to extend the reliability, modularity, and scalability of digital logic into processes for programmable assembly.
- Molecular Control and
Detection
Central to this effort is the ability to control and correct the placement of individual molecules. One approach being investigated is to attach metal nanoclusters to biological proteins that already do this, in order to be able to externally switch their conformation and hence function [K. Hamad-Schifferli, in Encyclopedia of Nanoscience and Nanotechnology, edited by J. A. Schwarz, C. Contescu and K. Putyera (Marcel Dekker, New York, 2003)]; a goal for the coming year is in vivo testing of RF biology. A major result over the last year has been extending this technique to demonstrate the attachment a superparamagnetic cobalt nanocluster to a DNA hairpin loop:
[B. Chow and J. Jacobson, to be submitted]. This introduces a resonant excitation in the RF spectrum, shown here as monitored by optical measurement of DNA dehybridization:
This high-Q tuning significantly extends the addressability of RF biology to the control of many more proteins simultaneously.
A second approach to molecular control being investigated is nanomechanics. CBA's Focused Ion Beam writer was used to notch an SOI substrate that was then fractured at that interface to create a controllable atomically-flat nanometer gap:
[A. Sprunt and A. Slocum, to be submitted]. This device will be investigated for use as a molecular-scale valve.
Along with molecular control, complementary work is being done on molecular detection. The field-effect sensor reported last year was successfully functionalized with DNA, to demonstrate an electronic readout of DNA hybridization [Electronic Detection of DNA by its Intrinsic Molecular Charge, J. Fritz, E.B. Cooper, S. Gaudet, et al., P. Nat. Acad. Sci (99) pp. 14142-14146 (2002)]. The charge per base at the sugar-phosphate backbone extends the size of an underlying depletion region in the device, shown here differentially detecting an oligonucleotide:
A goal for the coming year is to integrate microfluidics with MEMS biosensors.
- Self-Assembly
For molecular assembly to extend to the fabrication of macroscopic systems it must operate with massively parallel control of placement; for this reason work is being done in CBA on programming self-assembly over a number of length scales. A major result has been the development of the first amino acid-based surfactant nanomaterials, forming nanotubes and nanovesicles [Self-Assembly of Surfactant-Like Peptides with Variable Glycine Tails to Form Nanotubes and Nanovesicles, S. Santoso, W. Hwang, H. Hartman, S. Zhang, NanoLetters (2) pp. 687-691 (2002)] (left figure, below).
The possibility of patterning such molecular systems was introduced by a second result, showing the formation of a surface adlayer on a patterned self-assembled monolayer [Formation, Patterning, and Polymerization of Surface Adlayers Using Self-Assembled Monolayers as Templates, D.W. Mosley, M.A. Sellmyer, J. Jacobson, Proceedings of the Materials Research Society, 2003] (right figure, below). Ordered surface adlayers of bis-diacetylene molecule triaconta-10,12,19,21-tetraynoic acid amide (BisDA) were formed using amide hydrogen-bonding interactions with a pre-formed self-assembled monolayer of 16-mercaptohexadecanamide (16MHA) on gold, patterned by stamping with octadecanethiol. Previous reports of cross-linked monolayer sheets have involved formation of the monolayer first in Langmuir troughs; this system is formed directly on the substrate through hydrogen-bonded templated assembly.
Programmable self-assembly was also explored on meso- (um-mm) and macro- (mm-m) scales using CBA's rapid-prototyping tools. Mechanical finite state machines were constructed that self-assemble at the interface between two immiscible liquids (PDMS at a PFD/H20 interface; left figure, below) [S. Griffith and J. Jacobson, to be submitted], and three-dimensional tiles were developed along with supporting CAD tools to locally encode global shapes without requiring supporting external forms (right figure, below) [L. Sass, to be submitted].
- Patterning
Finally, direct writing was investigated as a way to create desired functions. On nanometer scales, CBA's Focused Ion Beam writer was used to show that it's possible to write wires with nanoparticles, shown here using carbon:
[Nanostructure Fabrication by Direct E-Beam Writing of Purely Inorganic Nanoparticles, D.S. Kong, V. Anant, A. Salomon, W. Delhagen, H. Nair, J. Varsanik and J.M. Jacobson, Intl. Conf. on Electron, Ion, and Photon Beam Technology and Nanofabrication, 2003]. Prior work in this area has required deposition from a gas phase; a goal for the coming year will be to demonstrate direct writing of simple circuits.
On longer length scales, CBA's rapid-prototyping tools were used in a compelling demonstration of patterning artificial materials, patterning arrays of composite wire and split-ring resonators to create a medium with a negative index of refraction:
[Experimental Observations of a Left-Handed Material That Obeys Snell's Law, A.A. Houck, J.B. Brock, I.L. Chuang, Phys. Rev. Lett. (90) 137401/1-4 (2003)]. This was the first measurement of the two-dimensional electromagnetic scattering profile for such a material, confirming theoretical predictions of a negative index (n = -0.36 in this case).
- "Statistical-Mechanical"
Engineering
The preceding projects point to the very real possibility of engineering on Avogadro scales. Managing complexity is already one of the most severe technological scaling constraints; efficiently and reliably organizing systems with enormous numbers of degrees of freedom is going to require a new kind of design practice. The development of statistical mechanics made it possible to make precise statements about the macroscopic properties of a system from a description of its microscopic constituents, without knowledge of its particular internal configuration. We now seek to not just describe but also direct behavior in this limit. CBA's effort to develop "statistical-mechanical" engineering is bringing together theoretical and experimental practice at the intersection of control theory, geometry, inference, device physics, and of course statistical mechanics itself.
Theoretically, a result from the last year was enlarging the applicability of semidefinite programming to large-scale distributed systems by generalizing the local connectivity to encompass discrete symmetry groups [Distributed Control of Systems over Discrete Groups, B. Recht and R. D'Andrea, submitted]. And experimentally, a focus has been on implementing graphical message-passing algorithms in analog circuits. This framework can reproduce much of traditional engineering practice while making a direct connection to device physics, taking advantage of degrees of freedom that are neglected by conventional digital logic. This may have immediate applications in improving speed, power consumption, and silicon area of circuits, and a much deeper consequence for being able to tape out enormously complex chips that can function with unreliable components. A family of "soft" gates is being developed that can route messages in field-programmable analog logic arrays (left figure, below), and these soft gates are being tested in spread-spectrum acquisition where they appear even in simulation to beat known algorithms for multi-terminal channel sharing [Low-Complexity LFSR Synchronization by Forward-Only Message Passing, B. Vigoda, J. Dauwels, N. Gershenfeld, and H.-A. Loeliger, to be submitted]. Related component development has included an ultra-low power medium-term analog storage cell that can operate on less than 50nW with a 500us sampling time and has been shown to achieve 11-bits of sampled precision (figure below, right) [M. O'Halloran and R. Sarpeshkar, to be published].
Probabilistic message-passing algorithms naturally extend to random "fungible" architectures, which are being tested with "pushpin" computers [Pushpin Computing System Overview: a Platform for Distributed, Embedded, Ubiquitous Sensor Networks, Joshua Lifton, Deva Seetharam, Michael Broxton, Joseph Paradiso, in F. Mattern and M. Naghshineh (eds): Pervasive 2002, Proceedings of the Pervasive Computing Conference, Zurich Switzerland, 26-28 August 2002, Springer Verlag, Berlin Heidelberg, pp. 139-151]. Over the last year a bare-die fungible pushpin computer has been developed and is currently being fabricated in order to reach systems sizes beyond the simulation limit [W. Butera, to be submitted], shown here passing self-organizing mobile code that discovers their geometry and then serves as an addressable display:
In the coming year, a goal is to extend message-passing algorithms to routing in distributed hierarchical networks, [J. Li, K. Sollins, to be submitted] seeking to identify local communications with global extremization of desired system properties [R. Krikorian, N. Gershenfeld, to be submitted].
The relationship between algorithms for inference on graphs and their physical embodiment is also being explored in CBA in vivo, including analysis of the neuronal basis of birdsong acquisition:
[S. Seung, to be submitted] and learning the dynamics of interacting communities [Learning Communities: Connectivity and Dynamics of Interacting Agents, T. Choudhury, B. Clarkson, S. Basu, and A. Pentland, Proceedings of the International Joint Conference on Neural Networks, 2003].
3. Describe the opportunities for training and development provided by your project.
- Student research
Last year CBA supported or contributed to the education of 48 graduate students across MIT, with 10 completing theses. There are 140 grad and undergrad students now trained to use CBA's rapid prototyping facilities; notable accomplishments include Terri Yu who receiving MIT's Orloff award for most meritorious service to the community, Jeff Brock receiving the award for the best Senior Physics Thesis (on the negative index materials), and Christopher Lyon winning the Ernst A. Guillemin Award for best Masters thesis in Electrical Engineering.
- Classes
CBA-related classes include:
- MAS.863: How to Make
(Almost) Anything
This enormously over-subscribed course teaches students with a wide range of backgrounds to use rapid-prototyping facilities supported by CBA
- MAS.862: The Physics of Information Technology
- MAS.864: The Nature
of Mathematical Modeling
- 8.371J, MAS.865J: Quantum
Information Science
Texts developed for these three classes led to the launch by Cambridge University Press of the Cambridge Series in Information and the Natural Sciences
- 8.13: Experimental
Physics
A quantum computing experiment was developed for MIT's Junior Laboratory, which has become the most popular module
- 7.86J, BE.481J, MAS.866J: Silicon Biology
- 6.151: Semiconductor
Devices Project Laboratory
Students in this project class successfully integrated microfluidics with microelectronics, leading to a new ongoing research program in this area
- 2.971: Second Summer
Program
Part of the Office of Minority Education’s program to assist minorities in adjusting to the MIT environment; student teams receive intensive training in the design process, and then practice that in challenging internships. - 2.993/2.996: Designing Paths to Peace
Teaches student teams advanced manufacturing techniques through the creation of personally-meaningful stone inlays that can be shared globally [Paths-to-Peace, A New Method for Teaching Design and Manufacturing, A. Slocum, S. Awtar, A. Elmouelhi, M. Graham, P. Willoughby, 2nd International Conference on Open Collaborative Design for Sustainable Innovation, Bangalore, India, 2002].
- Design That Matters
This student-run course teaches techniques for the design of appropriate advanced technologies; two student groups last year won awards in the IDEAS International Technology Prize competition.
- MAS.863: How to Make
(Almost) Anything
- Fab Labs
CBA's laboratory research on technologies for personal fabrication is complemented by the field "fab lab" program that is bringing prototype capabilities to under-served communities that have been beyond the reach of conventional technology development and deployment. Labs were opened in rural India and northern Norway, following initial sites in Boston and Costa Rica. By making accessible engineering in space (down to microns, through precision machining) and time (down to microseconds, through RISC microcontrollers), these facilities have been uncovering what can be thought of as instrumentation and fabrication divides, and suggesting that they can be addressed by bringing IT development rather than just IT to the masses [FAB LAB: An Alternate Model of ICT for Development, B. Mikhak, C. Lyon, T. Gorton, N. Gershenfeld, C. McEnnis, J. Taylor, 2nd International Conference on Open Collaborative Design for Sustainable Innovation, Bangalore, India, 2002].
- Education
CBA researchers have worked with the international network of Computer Clubhouses (with 80 centers in 15 countries) as an outreach and project test site for working with teenagers. A tabletop computer network modeling toolkit was developed for use there, allowing 10- to 12-year-old Clubhouse members to create and experiment with network structures; this system was also used in a trial with the United States Postal Service [Tabletop Process Modeling Toolkit: A Case Study in Modeling USPS Mailflow, T. Gorton, B. Mikhak, K. Paul, demonstration at CSCW 2002]. For still younger kids, a version of computerized "System Blocks" was developed to help children learn system dynamics concepts that were previously viewed as being too advanced [System Blocks: A Physical Interface for System Dynamics Learning, O. Zuckerman and M. Resnick, International System Dynamics Conference, New York City, 2003].
- Meetings
CBA has run internally and with partners a series of topical meetings in areas of interest to emerging communities:
- Science in Hollywood, Content and Communication, Los Angeles, January 2003
- Fab Labs,
Washington DC, May 2003
These two meetings were run with the National Academies' Office for the Public Understanding of Science, to explore new ways to engage people in science.
- Emergent Engineering,
MIT, October 2002
This brought together representatives from many different domains struggling with complexity scaling issues, including designers of chips, networks, and power grids. The meeting helped seed roadmaps for both research and support in this area.
- Embedded/Distributed IP,
MIT, July 2002
This meeting led to launch of the "Internet 0" industry initiative (described below).
- Development by Design,
Bangalore, India, December 2002
CBA contributed to this meeting that brought together the global community working on new technologies for sustainable development.
- Internet 0
I0 emerged from industrial interest in the embedded/distributed IP nodes that were being developed for use in CBA research; these are now growing into a set of standards for bringing IP connectivity to ultra-lightweight physical devices, based on 7 core principles:
- IP to leaf nodes
- Bit sizes larger than network sizes
- Shared (UWB) analog modulation
- Peers don't require servers
- Physical programming interfaces
- Compiled specification of layering
- Open standards
- Scanning Antiquities
Boston's Museum of Fine Arts has one of the world's greatest collections of unassembled fragments of antiquities. CBA's shop technician John Difrancesco collaborated with Drs. Florence Friedman (Brown University) and Walter Gilbert (Harvard University) to 3D scan, virtually assemble, and print them.
Products
In this section, you will be asked to describe the tangible products coming out of your project. Specifically:
1. What have you published as a result of this work?
- Journal publications
- Experimental Implementation of an Adiabatic Quantum Optimization Algorithm, M. Steffen, W. van Dam, T. Hogg, G. Breyta, and I. Chuang, Phys. Rev. Lett. (90) #067903 (2003)
- Physical One-Way Functions, R. Pappu, B. Recht, J. Taylor, N. Gershenfeld, Science (297) pp. 2026-2030 (2002)
- Power of Entanglement in Quantum Communication, S. Lloyd, Phys. Rev. Lett. (90) #167902 (2003)
- A Practical Architecture for Reliable Quantum Computers, M. Oskin, F.T. Chong, I.L. Chuang IL, Computer (35) 79-+ (2002)
- Electronic Detection of DNA by its Intrinsic Molecular Charge, J. Fritz, E.B. Cooper, S. Gaudet, et al., P. Nat. Acad. Sci (99) pp. 14142-14146 (2002)
- Self-Assembly of Surfactant-Like Peptides with Variable Glycine Tails to Form Nanotubes and Nanovesicles, S. Santoso, W. Hwang, H. Hartman, S. Zhang, NanoLetters (2) pp. 687-691 (2002)
- Formation, Patterning, and Polymerization of Surface Adlayers Using Self-Assembled Monolayers as Templates, D.W. Mosley, M.A. Sellmyer, J. Jacobson, Proceedings of the Materials Research Society, 2003
- Experimental Observations of a Left-Handed Material That Obeys Snell's Law, A.A. Houck, J.B. Brock, I.L. Chuang, Phys. Rev. Lett. (90) 137401/1-4 (2003)
- Learning Communities: Connectivity and Dynamics of Interacting Agents, T. Choudhury, B. Clarkson, S. Basu, and A. Pentland, Proceedings of the International Joint Conference on Neural Networks, 2003
- Pushpin Computing System Overview: a Platform for Distributed, Embedded, Ubiquitous Sensor Networks, Joshua Lifton, Deva Seetharam, Michael Broxton, Joseph Paradiso, in F. Mattern and M. Naghshineh (eds): Pervasive 2002, Proceedings of the Pervasive Computing Conference, Zurich Switzerland, 26-28 August 2002, Springer Verlag, Berlin Heidelberg, pp. 139-151
- Nanostructure Fabrication by Direct E-Beam Writing of Purely Inorganic Nanoparticles, D.S. Kong, V. Anant, A. Salomon, W. Delhagen, H. Nair, J. Varsanik and J.M. Jacobson, Intl. Conf. on Electron, Ion, and Photon Beam Technology and Nanofabrication, 2003
- System Blocks: A Physical Interface for System Dynamics Learning, O. Zuckerman and M. Resnick, International System Dynamics Conference, New York City, 2003
- FAB LAB: An Alternate Model of ICT for Development, B. Mikhak, C. Lyon, T. Gorton, N. Gershenfeld, C. McEnnis, J. Taylor, 2nd International Conference on Open Collaborative Design for Sustainable Innovation, Bangalore, India, 2002
- Paths-to-Peace, A New Method for Teaching Design and Manufacturing, A. Slocum, S. Awtar, A. Elmouelhi, M. Graham, P. Willoughby, 2nd International Conference on Open Collaborative Design for Sustainable Innovation, Bangalore, India, 2002
- Books or other non-periodical, one-time publications
- K. Hamad-Schifferli, in Encyclopedia
of Nanoscience and Nanotechnology, edited by J. A. Schwarz, C.
Contescu and K. Putyera (Marcel Dekker, New York, 2003)
- Sumit Basu
Ph.D. degree 2002 (EECS, MAS)
Thesis at: http://web.media.mit.edu/~sbasu/papers/thesis.pdf
Conversational Scene Analysis - Brian P. Clarkson
Ph.D. degree 2002 (MAS)
Life Patterns: structures from wearable sensors - Timothy Gorton
Masters degree 2003 (EECS, MAS)
Tangible Toolkits for Reflective Systems Modeling - Matthew Hancher
Masters degree 2003 (EECS, MAS)
A Motor Control Framework for Many-Axis Interactive Robots - Murali Kota
Ph.D. degree 2003 (MAS, EECS)
An Approach to Bridging Atom Optics and Bulk Spin Quatum Computation - Che King Leo
Masters degree 2002 (EECS, MAS)
Contact and Free-Gesture Tracking for Large Interactive Surfaces - Ji Li
Masters degree 2003 (EECS/LCS)
Improving Application-level Network Services with Regions - Yanni Loukissas
Masters degree 2003 (Architecture)
Rulebuilding: Exploring Design Worlds through End-Use Programming - Christopher Lyon
Masters degree 2003 (EECS, MAS)
Encouraging Innovation by Engineering the Learning Curve - Micah O’Halloran
Masters degree 2002 (EECS)
A Clock-Based Analog Memory Element for Integrated Circuits - Rahul Bhargava
Masters degree 2002 (MAS)
Designing a Computational Construction Kit for the Blind and Visusally Impaired - Jeff Brock
S.B. degree 2003 (Physics)
Refraction and Focusing in Negative Index Materials - Daniel Kornhauser
Masters degree 2002 (MAS)
Designing a Craft Computing Environment for Non-Industrial Settings
- Casey Smith
Masters degree 2002 (MAS)
Material Design for a Robotic Arts Studio - Matthias Steffen
Ph.D. degree 2003 (Stanford, MAS)
A Prototype Quantum Computer Using Nuclear Spins in Liquid Solution
- http://cba.mit.edu/
Primary CBA site
- http://cba.mit.edu/projects/fablab/
Support for fab labs
- http://cba.mit.edu/projects/I0/
Support for Internet 0
- http://www.media.mit.edu/physics/pedagogy/fab/
Support for "How To Make (almost) Anything"
- http://tower-support.media.mit.edu/
Support for use of the "Towers" system
3. What other specific products (databases, physical collections, educational aids, software, instruments, or the like) have you developed?
- Quantum computer controller
The need for generating complex pulse sequences to control quantum computing experiments in CBA led to the development of a gate-array-based real-time programmer, which is being released to the research community as open-source hardware and software.
- Distributed dense motor
controller
In support of CBA research on robotics and personal fabrication, a modular networked motor controller and supporting protocols were developed for operating dense arrays of actuators.
- Internet 0 prototypes
CBA has developed prototype Internet 0 devices, which are serving as reference designs for larger-scale trials.
- Towers
A modular computation and instrumentation assembly kit, including layers for A/D, D/A, embedded computing and networking, memory, sensors, and displays, along with an intuitive software development environment, designed to aid grass-roots invention.
- Fab lab tools
In support of the fab lab project, open-source graphical and modeling tools are being extended to enable CAD/CAM/CAE workflows in a fab lab setting.
