Modeling Kinematic Cellular Automata:
An Approach to Self Self-
Replication
Principal Investigator: Consultants:
Tiham...
Modeling Kinematic Cellular
Automata
zz Rationale
zz Benefits
zz Applications
zz Project Goals
zz Strategy
zz Accom...
Rationale
zz Why Self Self-Replication?
zz Why not Self Self-Assembly?
zz Why Kinematic Cellular
Automata?
zz Why bot...
Rationale: Why Self Self-Replication?
zz Revolutionary manufacturing
process
zz Nanotechnology
zz Massive reduction in...
Rationale: Why not SelfRationale: Self-- Assembly?Examples have been demonstratedExamples demonstratedBut…But… zzNot “Geno...
Rationale:
Why Kinematic Cellular Automata
(KCA)?
zz Combines Von Neumann’s two
designs
zz Increased flexibility
zz ...
Rationale:
Why Both Macro and Nano Scale?
zz Abstract design
zz Macro:
zz Possible with current technology
zz Useful ...
Benefit: Cost Reduction/LbBenefit: Lb11041 1041081012101610201024102106108WrenchPentiumPotatoLobsterEigler’sIBM AdDigitalW...
Benefit: Programmable MaterialsMaterialsSimple identical modules Simple zzFlow ModeFlow Mode zzPixelatedPixelatedMode Mode...
Application: Space
zz Exploration
zz Robust
zz Hyperflexible
zz Resource Utilization
zz Lower launch weight
zz Expan...
Project Goals
zz Characterize self self-replication
zz Quantify the complexity of Self Self-
Replicating System (SRS) m...
Project Strategy
zz Hybridize two self self-replication
models
zz Keep it simple
zz Make it complicated
zz Refine app...
Accomplishments
Goals Accomplishments
Characterize unexplored areaunexplored areaExplored MultiExplored Multi--Dimension...
Characterizing SelfCharacterizing Self--Replication: Adjusting the Freitas/MerkleFreitas/Merkle116116--Dimension Design Sp...
Quantifying Difficulty of SRS Design
1.00E+001.00E+011.00E+021.00E+031.00E+041.00E+051.00E+061.00E+071.00E+081.00E+09Unit...
Hierarchy
Computer
Biology
KCA SRS
Horse
Self-replicating System:
Useful
Processor
Brain and Muscles
Subsystems: ...
Original Approach: Feynman
method
1.1.Start with trivial selfStart self-- replication 2.2.Move the complexity out of the...
Original Approach: Feynman
method
““Plenty of room at the bottom”, topPlenty top--down, fractal approachapproachMystery
Refine Approach (by 180 180°)
••We should start at the
bottom level and work up
••Imitate Mother Nature
••The Trivial+...
The Bottom Bottom-up Approach
Well Well-ordered environment,
Simple inert parts
Symmetric facets
Modular cells
Assemb...
Subsystem Requirements
If atoms are analogous to bits,
then:
zz Memory/Bus -- --> Transporter
> zzMoves Parts
zzALU -...
Transporter Subsystem
(pink corner structural part)
Assembler Subsystem
(light blue preparation tool)
(yellow edge structural part)
(pink corner structural part)
Controller Subsystem
Cell Design Requirements
zz Structure:
zz Lock, 1 1-D slide, disconnect
zz Actuators:
zz Transform
zz Move
zz Sensor...
Unit Cell
(center structure, motors, sensors, and tabs omitted)
Facet Design Requirements
zz Structure:
zz Insert or retract
zz Actuators:
zz Transform
zz Move tabs
zz Sensors:
zz...
Unit Facets
Parts Design Requirements
zz Structure:
zz Solid
zz Motors:
zz Rotary
zz Linear
zz IMPC
zz Sensors:
zz Translate s...
Parts: Structure, Sensors &
Solenoids
Software Simulation
zz Sensor Position Simulation Tool
zz NAND gate & op op-amp Self Self-Assembly
Tool
zz Facet Anima...
Position Sensor Simulation
Self Self-assembly of
NAND gate and op op-amp
Facet Animation
Simulation of
Transporter and Assembler
Conclusion and Future
Directions
No roadblocks!
zz Final Design for macro physical prototypes
zz Build physical protot...
Acknowledgements
zz NASA Institute for Advanced Concepts
zz John Sauter –– Altarum
zz Rick Berthiaume, Ed Waltz, Ken Au...
Additional Material
zz Assumptions
zz Previous and Related Work
KCA SRS Assumptions
zz Parts supplied as automated cartridges
zz Low rate of errors detected in code
Previous and Related Work
zz Freitas and Long - NASA Summer Study:
Advanced Automation for Space Missions (1980)
zz Mic...
Previous Work:NASA Summer Study
Advanced Automation for Space
Missions - Freitas and Long -
(1980)
zz Strengths
zz Fi...
Previous Work: Joseph
Michael
zzStrengths zz“The DOS of Utility Fog”“Fog” zzWorking macro modular robotsrobots zzLimited...
Previous Work: Forrest BishopBishop zzStrengths zzVery Limited DOFVery DOF zzClear macro designClear design zzWeaknessesWe...
Related Work:
Chirikjian/Suthakorn
zzStrengthsStrengths zzAutonomous implementation of Trivialof Trivial+2+2case (4 part...
Related Work: Zyvex
zzProjectsProjects zzApplying MEMS and nanotubesnanotubes zzParallel Micro and Exponential AssemblyAs...
Related Work: Freitas/Merkle
Kinematic SelfKinematic Self--Replicating Machines ((LandesLandesBioscience, 2004)Bioscience...
Related Work: Matt Moses
zStrengths:
zCAD to physical implementation
zLarge envelope UC
zWeaknesses: zStrain bending u...
Why Universal Constructors?Envelope = everythingRockUCEnvelope = nothingAssembly LineEnvelope = one thingRobotEnvelope = m...
of 48

Kinematic cellularautomata

Published on: Mar 3, 2016
Source: www.slideshare.net


Transcripts - Kinematic cellularautomata

  • 1. Modeling Kinematic Cellular Automata: An Approach to Self Self- Replication Principal Investigator: Consultants: Tihamer Toth Toth-Fejel Robert Freitas Tihamer.Toth Toth-Fejel@gd gd-ais.com Matt Moses March 22, 2004 NIAC Fellows Presentation NASA Institute for Advanced Concepts Phase I: CP CP-02 02-02
  • 2. Modeling Kinematic Cellular Automata zz Rationale zz Benefits zz Applications zz Project Goals zz Strategy zz Accomplishments zz Conclusion and Future Directions zz Additional Material
  • 3. Rationale zz Why Self Self-Replication? zz Why not Self Self-Assembly? zz Why Kinematic Cellular Automata? zz Why both macro and nano scale?
  • 4. Rationale: Why Self Self-Replication? zz Revolutionary manufacturing process zz Nanotechnology zz Massive reduction in costs per pound zz Controlled exponential growth
  • 5. Rationale: Why not SelfRationale: Self-- Assembly?Examples have been demonstratedExamples demonstratedBut…But… zzNot “Genotype + RibotypeRibotype= Phenotype” (GRP)= GRP) zzNo theoryNo theory zzAgainst the principles of sound designAgainst designHowever…However… Use it for simple input partsUse parts
  • 6. Rationale: Why Kinematic Cellular Automata (KCA)? zz Combines Von Neumann’s two designs zz Increased flexibility zz Decreased complexity zz Large system work envelope zz Sometimes better than smart dust
  • 7. Rationale: Why Both Macro and Nano Scale? zz Abstract design zz Macro: zz Possible with current technology zz Useful products zz Proof of concept in short term zz Nano Nano: zz Quality of atoms (and molecules) zz Self Self-assembled input parts possible zz Significant financial payoff
  • 8. Benefit: Cost Reduction/LbBenefit: Lb11041 1041081012101610201024102106108WrenchPentiumPotatoLobsterEigler’sIBM AdDigitalWatchCarYogurtSaltKCA SRSComplexity (Parts and Interactions/lb) Cost ($/lb) SevenSevenMagnitudes!Magnitudes! GAIN Traditional Top Down ManufacturingvsBottom-up Molecular Replication
  • 9. Benefit: Programmable MaterialsMaterialsSimple identical modules Simple zzFlow ModeFlow Mode zzPixelatedPixelatedMode Mode zzLogic Processing ModeLogic ModeFlow ModeFlow ModePixelatedPixelatedMode Mode
  • 10. Application: Space zz Exploration zz Robust zz Hyperflexible zz Resource Utilization zz Lower launch weight zz Expandable zz Terraforming zz Politically feasible zz Opens new frontier
  • 11. Project Goals zz Characterize self self-replication zz Quantify the complexity of Self Self- Replicating System (SRS) made of Kinematic Cellular Automata (KCA) zz Confirm approach zz Design a KCA SRS zz Simulate designs
  • 12. Project Strategy zz Hybridize two self self-replication models zz Keep it simple zz Make it complicated zz Refine approach zz Attempt design zz Imitate computers zz Imitate biology
  • 13. Accomplishments Goals Accomplishments Characterize unexplored areaunexplored areaExplored MultiExplored Multi--Dimensional SpaceDimensional SpaceQuantify the Quantify difficultydifficultyNot trivial, but less than a PentiumNot PentiumConfirm or refute Confirm approach Refined ApproachRefined Approach zzUseful SRSUseful SRS zzHierarchy of Subsystems, Cells, Facets, & PartsHierarchy Parts zzTransporter, Assembler, & ControllerTransporter, Controller zzLowLow--level simpler than highlevel high--levellevel zzTopTop--Down vsvsBottomBottom--UpUp zzSelfSelf--Assembly for input PartsAssembly Parts zzStandard conceptsStandard concepts zzUniversal Constructor is approach, not goalUniversal goalDesign a KCA SRSDesign SRSDeveloped Requirements Developed Preliminary DesignPreliminary DesignSimulate designsSimulate designsModeled SimulationsModeled Simulations zzSensor PositionSensor Position zzNAND gate and opNAND op--amp selfamp self--assemblyassembly zzFacetFacet zzTransporter and AssemblerTransporter Assembler
  • 14. Characterizing SelfCharacterizing Self--Replication: Adjusting the Freitas/MerkleFreitas/Merkle116116--Dimension Design SpaceSpaceEvolvabilityReplicatorInformationReplicatorEnergeticsReplicationProcess Product Structure Replicator Performance Replicator Substrate Replicator Control Manipulation AutonomyManipulationRedundancyManipulationCentralizationManipulationDegrees of FreedomPositionalAccuracy Quantity of Onboard Energy Types Quantity of Manipulation Types Assembly StyleParts CountParts ScalePartsTypesQuantityof VitaminPartsNutritionalComplexityPartsPreparationPartsComplexityPartsPrecisionMulticellularity Subunit Scale Subunit Types Subunit ComplexityReplicatorKinematicsActiveSubunitsReplicatorPartsAllReplicatorsReplicatorStructureSubunitHierarchy Replicator Designability Subunit Complexity Monotonicity
  • 15. Quantifying Difficulty of SRS Design 1.00E+001.00E+011.00E+021.00E+031.00E+041.00E+051.00E+061.00E+071.00E+081.00E+09Units of difficulty# parts+ # interactionsWrenchAutomobile Pentium KCA SRS
  • 16. Hierarchy Computer Biology KCA SRS Horse Self-replicating System: Useful Processor Brain and Muscles Subsystems: Transporter, Assembler, and Controller Bus/Memory, ALU, and Controller Cells Cells: Cubic devices with only three limited degrees of freedom Organelles Facets: Symmetrical implementation Finite State Machines, Shift Registers, Adders, and Multiplexers Proteins Parts: Inert, Simpler than higher levels NAND gates Genes Self-assembling Subparts: Wires, Transistors, actuator components Transistors, Wires Molecules Molecules Molecules
  • 17. Original Approach: Feynman method 1.1.Start with trivial selfStart self-- replication 2.2.Move the complexity out of the environment and into the SRS by doubling parts count of the component (Trivialthe Trivial+1+1case)case) 3.3.ReiterateReiterate
  • 18. Original Approach: Feynman method ““Plenty of room at the bottom”, topPlenty top--down, fractal approachapproachMystery
  • 19. Refine Approach (by 180 180°) ••We should start at the bottom level and work up ••Imitate Mother Nature ••The Trivial+2 case has already been done Molecules
  • 20. The Bottom Bottom-up Approach Well Well-ordered environment, Simple inert parts Symmetric facets Modular cells Assembler, Transporter, and Controller subsystems Self Self-Replicating System
  • 21. Subsystem Requirements If atoms are analogous to bits, then: zz Memory/Bus -- --> Transporter > zzMoves Parts zzALU -- --> Assembler > zzConnects Parts zz Control -- --> Controller > zzDecides which Parts go where zzStandardized
  • 22. Transporter Subsystem (pink corner structural part)
  • 23. Assembler Subsystem (light blue preparation tool) (yellow edge structural part) (pink corner structural part)
  • 24. Controller Subsystem
  • 25. Cell Design Requirements zz Structure: zz Lock, 1 1-D slide, disconnect zz Actuators: zz Transform zz Move zz Sensors: zz Detect Position zz Transmit messages zz Logic: zz Decode messages zz Accept, store, forward messages zz Activate commands
  • 26. Unit Cell (center structure, motors, sensors, and tabs omitted)
  • 27. Facet Design Requirements zz Structure: zz Insert or retract zz Actuators: zz Transform zz Move tabs zz Sensors: zz Transform zz Logic: zz Decode
  • 28. Unit Facets
  • 29. Parts Design Requirements zz Structure: zz Solid zz Motors: zz Rotary zz Linear zz IMPC zz Sensors: zz Translate signals zz Detect parts position zz Logic zz Activate messages to motors zz Aggregate digital logic
  • 30. Parts: Structure, Sensors & Solenoids
  • 31. Software Simulation zz Sensor Position Simulation Tool zz NAND gate & op op-amp Self Self-Assembly Tool zz Facet Animation zz Transporter and Assembler Simulation
  • 32. Position Sensor Simulation
  • 33. Self Self-assembly of NAND gate and op op-amp
  • 34. Facet Animation
  • 35. Simulation of Transporter and Assembler
  • 36. Conclusion and Future Directions No roadblocks! zz Final Design for macro physical prototypes zz Build physical prototypes zz Build and run small cell collections zz Build and run subsystems zz Build macro scale SRS zz Write Place and Route software zz Refine concept at nano scale
  • 37. Acknowledgements zz NASA Institute for Advanced Concepts zz John Sauter –– Altarum zz Rick Berthiaume, Ed Waltz, Ken Augustyn, and Sherwood Spring –– General Dynamics AIS zz John McMillan and Teresa Macaulay –– Wise Solutions zz Forrest Bishop –– Institute of Atomic Atomic-Scale Engineering zz Joseph Michael –– Fractal Robots, Ltd.
  • 38. Additional Material zz Assumptions zz Previous and Related Work
  • 39. KCA SRS Assumptions zz Parts supplied as automated cartridges zz Low rate of errors detected in code
  • 40. Previous and Related Work zz Freitas and Long - NASA Summer Study: Advanced Automation for Space Missions (1980) zz Michael - Fractal Robots zz Chirikjian and Suthakorn - Autonomous Robots zz Moses - Universal Constructor Prototype zz Zyvex - Exponential Assemblers zz Freitas and Merkle - Kinematic Self Self-Replicating Machines (2004)
  • 41. Previous Work:NASA Summer Study Advanced Automation for Space Missions - Freitas and Long - (1980) zz Strengths zz First major work since 1950s zz Cooperation of many visionaries zz Weaknesses zz short, no follow follow-up zz paper study only zz pre pre-PC technology http://www.islandone.org/MMSG/aasm/ FOR MORE INFO...
  • 42. Previous Work: Joseph Michael zzStrengths zz“The DOS of Utility Fog”“Fog” zzWorking macro modular robotsrobots zzLimited DOF = better structurestructure zzWeaknessesWeaknesses zzFractals just push problem to lower, lesslower, less--accessible levelaccessible level zzno detailed methodology for selfself--replicationreplicationhttp://www.fractal-robots.com/ FOR MORE INFO...
  • 43. Previous Work: Forrest BishopBishop zzStrengths zzVery Limited DOFVery DOF zzClear macro designClear design zzWeaknessesWeaknesses zzNanoscaleNanoscaleimplementation clearly implementation implied, but not clearly designeddesigned zzno detailed methodology for selffor self--replicationreplicationhttp://www.iase.cc/html/overtool.htmFOR MORE INFO...
  • 44. Related Work: Chirikjian/Suthakorn zzStrengthsStrengths zzAutonomous implementation of Trivialof Trivial+2+2case (4 parts)case zzDirected towards extraterrestrial applicationsextraterrestrial applications zzLego isomorphic with moleculesmolecules zzWeaknessesWeaknesses zzSmall UC envelopeSmall envelope zzDepends on nonDepends non--replicating jigsjigs zzHigh entropy environment limits extension to Triviallimits Trivial+3+3case case http://caesar.me.jhu.edu/research/self_replicating.htmlcaesar.htmlFOR MORE INFO...
  • 45. Related Work: Zyvex zzProjectsProjects zzApplying MEMS and nanotubesnanotubes zzParallel Micro and Exponential AssemblyAssembly zzStrengthsStrengths zzFirst and only funded company trying to build a DrexlerianDrexlerianassemblerassembler zzWeaknessesWeaknesses zzMEMS is 1000X too bigMEMS big zzsurfaces too roughsurfaces rough zzExponential Assembly is machine selfExponential self-- assembly (not Universal Constructor; not GRP paradigm; not Utility Fog)FOR MORE INFO... http:// www.zyvex.com com/
  • 46. Related Work: Freitas/Merkle Kinematic SelfKinematic Self--Replicating Machines ((LandesLandesBioscience, 2004)Bioscience, First comprehensive review of fieldFirst field1.1.The Concept of SelfThe Self--Replicating MachinesMachines2.2.Classical Theory of Machine ReplicationClassical Replication3.3.MacroscaleMacroscaleKinematic Machine Kinematic ReplicatorsReplicators4.4.MicroscaleMicroscaleand Molecular Kinematic and Machine ReplicatorsMachine Replicators5.5.Issues in Kinematic Machine Replication EngineeringEngineering6.6.Motivations for MolecularMotivations Molecular--Scale Machine Replicator DesignReplicator Design(c) 2004 Robert Freitas and Ralph MerkleFreitas is a technical consultant for this project
  • 47. Related Work: Matt Moses zStrengths: zCAD to physical implementation zLarge envelope UC zWeaknesses: zStrain bending under load zManual controlMoses is a technical consultant for this project
  • 48. Why Universal Constructors?Envelope = everythingRockUCEnvelope = nothingAssembly LineEnvelope = one thingRobotEnvelope = many thingsSRSEnvelope = every constituent part zzUC is the ability to build anythingUC anything zzUses “Genotype+RibotypeGenotype+Ribotype= Phenotype”= Phenotype” zzConstruction envelope includes itselfConstruction itself zzAtoms equivalent to bitsAtoms bits zzSRS only needs limited UC capabilitySRS capability

Related Documents