Dran Computer Science Lab

Byoungkwon An (Kwon An)

(Byoungkwon An) — Thu, 20 Dec 2012 05:55:22 -0500

Byoungkwon An ( Kwon An )

e-mail: dran at csail.mit.edu
web: www.drancom.com

Resume

*. Dran Computer Science Lab is An's private Lab for his independent research. He established the lab with his MSX computer when he was 6 years old. (since 1987)

*. Some projects have been completed under Prof. Daniela Rus's guidence in CSAIL at MIT.

Projects

Print-Like Manufacturing System for Smart Sheets

Distributed Computing for Self-Assembling Modular Systems

Distributed Algorithms for Self-Reconfiguring Sliding Cube Modular System

Computation Theory for Many-Core Processors

Previous Projects

Programmable Smart Sheet (CSAIL, MIT)

Folding Planning Algorithms for Smart Sheets (CSAIL, MIT)

Distributed Shape Magnification Algorithm for Self-Assembling Modular Systems (CSAIL, MIT)

EM-Cube 3 : Cube-shaped Self-Reconfigurable Robots

TF-series : Micromachined Cube-shaped Self-Reconfigurable Robots

ME-Cube 2 : Cube-shaped Self-Reconfigurable Robots

EM-Cube 1 : Cube-shaped Self-Reconfigurable Robots

Distributed-Table-based Artificial Nervous System

Field Effect Transistor Nerve-Computer Interface (Patent:10-0765960, KR)

Atomic Force Microscope (AFM)
(Nanophysics Lab., Korea University)

e-DNA and Stem Cell Algorithm : Transfomation Algorithm for >1000 microscale regular solid shaped Self-Reconfigurable Robots

Descriptions for Projects (pdf)

Journal Papers

Byoungkwon An, Daniela Rus, Programming and Designing Self-Folding Sheets, Robotics and Autonomous Systems, Submitted, Oct. 2012 (under Review)

Byoungkwon An, Nadia Benbernou, Erik Demaine, Daniela Rus, Planning to Fold Multiple Objects from a Single Self-Folding Sheet, Robotica, 29(1): 87-102, 2011

Elliot Hawkes, Byoungkwon An, Nadia Benbernou, Hiroto Tanaka, Sangbae Kim, Erik Demaine, Daniela Rus, Robert Wood, Programmable Matter by Folding, Proceedings of the National Academy of Sciences (PNAS), 107(28): 12441-12445, 2010 url pdf

Conference Papers

Byoungkwon An, Daniela Rus, Programming and Controlling Self-Folding Robots, IEEE International Conference on Robotics and Automation (ICRA), St. Paul, Minnesota, 2012

Jamie Paik, Byoungkwon An, Daniela Rus, Robert Wood, Robotic Origamis: Self-Morphing Modular Robots, International Conference on Morphological Computation (ICMC), Venice, Italy, 2011

Byoungkwon An, Daniela Rus, Making Shapes from Modules by Magnification, IEEE/RSJ International Conference on Intelligent Robots and System (IROS), Taipei, Taiwan, 2010 pdf

Byoungkwon An, EM-Cube: Cube-shaped, Self-Reconfigurable Robots Sliding on Structure Surface, IEEE International Conference on Robotics and Automation (ICRA), Pasadena, CA, 2008 pdf

Thesis

Sticker Controller and Sticker Programming for Smart Sheets (Self-Folding Sheets), Master of Science, Computer Science, MIT, September 2011

Scaling of Dynamic Surface Growth Model on Fractal Dimension, Bachelor of Science, Physics, Soongsil University, February 2004

Patent

ByoungKwon An, FET (Field Effect Transistor) Nerve Electronic Chip, 10-0765960, KR

Media Coverage (Selective)

CNN, Edge of Discovery, Transformers could be a reality!, June 1, 2011

MIT News, Shape-shifting robots, August 4, 2010

Discovery News, Origami Robot Makes Shapes on Demand, July 7, 2010

CNET, Robotics meet origami in self-folding sheets, June 29, 2010

Nature News, Origami that folds itself, June 28, 2010

MSNBC, 'Programmable matter' may shape future tools, June 28, 2010

Scientific American, Shifty Science: Programmable Matter Takes Shape with Self-Folding Origami Sheets, June 28, 2010

Popular Science, Video: "Smart Sheets" Can Self-Assemble Into Airplanes, Boats, June 28, 2010

Phys.org, Shape-shifting sheets automatically fold into multiple shapes, June 28, 2010

Harvard Gazette, Shape-shifting sheets automatically fold into multiple shapes, June 28, 2010

Snowboard

Certified Snowboard Instructor, Canadian Association of Snowboard Instructors (Level 1)

(last update : Jan. 3, 2013)

Distributed Computing for Self-Assembling Modular System

(Byoungkwon An) — Thu, 20 Dec 2012 05:54:06 -0500

Distributed Computing for Self-Assembling Modular System

Table of Contents
1. Self-Assembling Modular System
2. Distributed Shape Magnification Algorithms
3. Processor Architecture for Self-Assembling Modular Systems (Current)
4. Theoretical Computation Machine for Processor of Self-Assembling Modular Systems (Current)
5. References

1. Self-Assembling Modular System
A self-assembling modular system is a distributed system composed of many robotic modules. By attaching selected modules to the neihbor modules, the system assembles a object shape itself. . Each robotic module contains connecters to attach the selected neighborhood modules as well as a micro-processor to communicatate with the neighborhood modules and to control the connecters. Smart pabble robots [2] or Miche [3] are the examples of self-assembling modular systems.

Video 1, 2: Example of self-assembling modular systems. Smart pabble robots [4,2].

2. Distributed Shape Magnification Algorithms [1]

A distributed shape magnification algorithm is a distributed algorithm that parallelly controls a self-assembling modular system to connect the selected modules for the given target objects [1]. Each module does not have any inforamtion of its id and location. Although each module just locally repeats the algorithm when it receives the messages, the whole system complites assembling the target objects. (Video 3).

Details of the algorithms, proofs and simulation resultes are explained in [1]. A example of magnification of "h" shape is shown as in Fig. 4, 5.

Video 3: Simulation of Making Shape by Distrubuted Shape Magnificaiton Algorithm [1]

" width="400" height="300">

[Flash]

Fig. 4. Example of input "h" shape coordinates into the seed module [1].
alg_input

Fig. 5. An example of magnification "h" shape in 2D. Magnification factors are 1, 2, and 3 from left to right figures. Top figures are the self-assembling modules in the bag. Bottom figures are the connected the modules which is the magnified desired shape. The red module is the seed module. The modules on green are marked itself as internal connecting module. The modules on blue are disconnecting modules.
alg_mag

3. Processor Architecture for Self-Assembling Modular Systems (Current)
We explore the processor of each modula to understand the minimum requriements of the instruction set and the registors for the algorithms. We built and proved a main algorithm and its extention algorithms for 3D self-assembling system. An ISA(instruction set architecture) composed of 5 registors, 5 instructions with no jump instruction can run the main algorithm. We are designing the logic layer of this ISA, to miximize the parallal computation of each processor.

4. Theoretical Computation Machine for Processor of Self-Assembling Modular Systems (Current)
We explore the computation in theoretical level by designing and analyzing 2D-tape distributed-head turing machine. Although the machine contains 7-tuple like a turning machine, it contains a different tape movement and a transition function for parallel computation.

5. References
[1] An, B., Rus, D., Making Shapes from Modules by Magnification, IEEE/RSJ International Conference on Intelligent Robots and System (IROS), 2010, url
[2] Gilpin, K., Rus, D., A Distributed Algorithm for 2D Shape Duplication with Smart Pebble Robots, IEEE International Conference on Robotics and Automation (ICRA), 2012
[3] Gilpin, K., Kotay, K., Rus, D., V. Iuliu, Miche: Modular Shape Formation by Self-Disassembly, The International Journal of Robotics Research, 27(3-4):345-372, 2008
[4] http://newsdirect.nma.com.tw/SingleItem.aspx?asset_id=OEM_20120408_OINT_002

Cube-shape Generic Distributed Self-Reconfigurable Robots

(Byoungkwon An) — Sat, 30 Dec 2006 11:47:12 -0500

Cube-shape Self-Reconfigurable Robots
Description for project, pdf

< EM-Cube 3 >

Description for project, pdf

Locomotion video (400kb,DivX)

< Microscale TF series >

Description for project, pdf

<ME-Cube 2>

Description for project, pdf

<EM-Cube 1>

Description for project, pdf

(last updated : 2007. 07. 28)

Distributed-Table-based (DTB) Artificial Nervous System

(Byoungkwon An) — Sat, 30 Dec 2006 11:47:02 -0500

Distributed-Table-based (DTB)
Artificial Nervous System

One-page Description (pdf)

(last updated : 2007. 7. 28)

Field Effect Transistor Nerve-Computer Interface

(Byoungkwon An) — Sat, 30 Dec 2006 11:13:43 -0500

Field Effect Transistor Nerve-Computer Interface

One-page Description, pdf

(last updated: 2007. 7. 28)