Tihamer Toth-Fejel - NASA's Institute for Advanced Concepts
Tihamer Toth-Fejel - NASA's Institute for Advanced Concepts
Tihamer Toth-Fejel - NASA's Institute for Advanced Concepts
- No tags were found...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Modeling Kinematic CellularAutomata• Rationale• Benefits• Applications• Project Goals• Strategy• Accomplishments• Conclusion and FutureDirections• Additional Material
Rationale• Why Self-Replication?• Why not Self-Assembly?• Why Kinematic CellularAutomata?• Why both macro and nano scale?
Rationale: Why Self-Replication?• Revolutionary manufacturingprocess• Nanotechnology• Massive reduction in costs perpound• Controlled exponential growth
Rationale: Why not Self-Assembly?Examples have been demonstratedBut…• Not “Genotype + Ribotype = Phenotype” (GRP)• No theory• Against the principles of sound designHowever…Use it <strong>for</strong> simple input parts
Rationale:Why Kinematic Cellular Automata(KCA)?• Combines Von Neumann’s twodesigns• Increased flexibility• Decreased complexity• Large system work envelope• Sometimes better than smart dust
Rationale:Why Both Macro and Nano Scale?• Abstract design• Macro:• Possible with current technology• Useful products• Proof of concept in short term• Nano:• Quality of atoms (and molecules)• Self-assembled input parts possible• Significant financial payoff
Benefit: ProgrammableMaterialsSimple identical modules• Flow Mode• Pixelated Mode• Logic Processing ModeFlow ModePixelated Mode
Application: Space• Exploration• Robust• Hyperflexible• Resource Utilization• Lower launch weight• Expandable• Terra<strong>for</strong>ming• Politically feasible• Opens new frontier
Project Goals• Characterize self-replication• Quantify the complexity of Self-Replicating System (SRS) madeof Kinematic Cellular Automata(KCA)• Confirm approach• Design a KCA SRS• Simulate designs
Project Strategy• Hybridize two self-replicationmodels• Keep it simple• Make it complicated• Refine approach• Attempt design• Imitate computers• Imitate biology
AccomplishmentsGoalsCharacterizeunexplored areaQuantify thedifficultyConfirm or refuteapproachDesign a KCA SRSSimulate designsAccomplishmentsExplored Multi-Dimensional SpaceNot trivial, but less than a PentiumRefined Approach• Useful SRS• Hierarchy of Subsystems, Cells, Facets, & Parts• Transporter, Assembler, & Controller• Low-level simpler than high-level• Top-Downvs Bottom-Up• Self-Assembly <strong>for</strong> input Parts• Standard concepts• Universal Constructor is approach, not goalDeveloped RequirementsPreliminary DesignModeled Simulations• Sensor Position• NAND gate and op-amp self-assembly• Facet• Transporter and Assembler
Characterizing Self-Replication: Adjustingthe Freitas/Merkle 116-Dimension DesignSpaceReplicatorSubstrateReplicatorControlReplicatorDesignabilityReplicatorPer<strong>for</strong>manceReplicatorIn<strong>for</strong>mationProductStructureAllReplicatorsEvolvabilityReplicationProcessReplicatorStructureReplicatorEnergeticsPositionalAccuracyAssemblyStylePartsCountPartsScaleReplicatorPartsManipulationDegrees of FreedomPartsTypesMulticellularityActiveSubunitsSubunitHierarchyReplicatorKinematicsQuantity ofManipulationTypesQuantityof VitaminPartsNutritionalComplexityManipulationAutonomyManipulationRedundancyPartsPreparationPartsComplexityPartsPrecisionSubunitScaleSubunitComplexitySubunitTypesSubunitComplexityMonotonicityManipulationCentralizationQuantity of OnboardEnergy Types
Quantifying Difficulty of SRS DesignUnits ofdifficulty# parts+# interactions1.00E+091.00E+081.00E+071.00E+061.00E+051.00E+041.00E+031.00E+021.00E+011.00E+00Wrench Automobile Pentium KCA SRS
HierarchyBiologyHorseBrain andMusclesCellsOrganellesKCA SRSSelf-replicating System:UsefulSubsystems:Transporter, Assembler, and ControllerCells: Cubic devices with only threelimited degrees of freedomFacets: Symmetrical implementationComputerProcessorBus/Memory, ALU,and ControllerFinite StateMachines,Shift Registers,Adders, andMultiplexersProteins Parts: Inert, Simpler than higher levels NAND gatesGenesSelf-assembling Subparts:Wires, Transistors, actuator componentsTransistors, WiresMolecules Molecules Molecules
Original Approach: Feynmanmethod1. Start with trivial self-replication2. Move the complexity outof the environment andinto the SRS bydoubling parts count ofthe component (Trivial +1case)3. Reiterate
Original Approach: Feynmanmethod“Plenty of room at the bottom”, top-down, fractalapproachMystery
Refine Approach (by 180°)•We should start at thebottom level and work up•Imitate Mother Nature•The Trivial+2 case hasalready been doneMolecules
The Bottom-up ApproachWell-ordered environment,Simple inert partsSymmetric facetsModular cellsAssembler, Transporter, and ControllersubsystemsSelf-ReplicatingSystem
Subsystem RequirementsIf atoms are analogous to bits,then:• Memory/Bus --> > Transporter•Moves Moves Parts• ALU --> > Assembler•Connects Parts• Control --> > Controller•Decides Decides which Parts go where•Standardized
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• Structure:• Lock, 1-D 1 D slide, disconnect• Actuators:• Trans<strong>for</strong>m• Move• Sensors:• Detect Position• Transmit messages• Logic:• Decode messages• Accept, store, <strong>for</strong>ward messages• Activate commands
Unit Cell(center structure, motors, sensors, and tabs omitted)
Unit Facets
Parts Design Requirements• Structure:• Solid• Motors:• Rotary• Linear• IMPC• Sensors:• Translate signals• Detect parts position• Logic• Activate messages to motors• Aggregate digital logic
Parts: Structure, Sensors &Solenoids
Software Simulation• Sensor Position Simulation Tool• NAND gate & op-amp Self-AssemblyTool• Facet Animation• Transporter and Assembler Simulation
Position Sensor Simulation
Self-assembly ofNAND gate and op-amp
Facet Animation
Simulation ofTransporter and Assembler
Conclusion and FutureNo roadblocks!Directions• Final Design <strong>for</strong> macro physical prototypes• Build physical prototypes• Build and run small cell collections• Build and run subsystems• Build macro scale SRS• Write Place and Route software• Refine concept at nano scale
Acknowledgements• NASA <strong>Institute</strong> <strong>for</strong> <strong>Advanced</strong> <strong>Concepts</strong>• John Sauter – Altarum• Rick Berthiaume, Ed Waltz, Ken Augustyn, andSherwood Spring – General Dynamics AIS• John McMillan and Teresa Macaulay– Wise Solutions• Forrest Bishop– <strong>Institute</strong> of Atomic-Scale Engineering• Joseph Michael – Fractal Robots, Ltd.
Additional Material• Assumptions• Previous and Related Work
KCA SRS Assumptions• Parts supplied as automated cartridges• Low rate of errors detected in code
Previous and Related Work• Freitas and Long - NASA Summer Study:<strong>Advanced</strong> Automation <strong>for</strong> Space Missions (1980)• Michael - Fractal Robots• Chirikjian and Suthakorn - Autonomous Robots• Moses - Universal Constructor Prototype• Zyvex - Exponential Assemblers• Freitas and Merkle - Kinematic Self-ReplicatingMachines (2004)
Previous Work:NASA Summer StudyFOR MORE INFO...http://www.islandone.org/MMSG/aasm/<strong>Advanced</strong> Automation <strong>for</strong> SpaceMissions - Freitas and Long -(1980)• Strengths• First major work since 1950s• Cooperation of manyvisionaries• Weaknesses• short, no follow-up• paper study only• pre-PC technology
Previous Work: JosephFOR MORE INFO...http://www.fractal-robots.com/Michael• Strengths• “The DOS of Utility Fog”• Working macro modularrobots• Limited DOF = betterstructure• Weaknesses• Fractals just push problem tolower, less-accessible level• no detailed methodology <strong>for</strong>self-replication
Previous Work: Forrest• Strengths• Very Limited DOF• Clear macro design• Weaknesses• Nanoscaleimplementation clearlyimplied, but not clearlydesigned• no detailed methodology<strong>for</strong> self-replicationBishopFOR MORE INFO...http://www.iase.cc/html/overtool.htm
Related Work:Chirikjian/Suthakorn• Strengths• Autonomous implementationof Trivial +2 case (4 parts)• Directed towardsextraterrestrial applications• Lego isomorphic withmolecules• Weaknesses• Small UC envelope• Depends on non-replicatingjigs• High entropy environmentFOR MORE INFO...limits extension to Trivial +3http://caesar.me.jhu.edu/research/self_replicating.htmlcase
Related Work: ZyvexFOR MORE INFO...• Projects• Applying MEMS and nanotubes• Parallel Micro and ExponentialAssembly• Strengths• First and only funded company tryingto build a Drexlerian assembler• Weaknesses• MEMS is 1000X too big• surfaces too rough• Exponential Assembly is machine self-assembly (not Universal Constructor;not GRP paradigm; not Utility Fog)http://www.zyvex.comwww.zyvex.com/
Related Work: Freitas/MerkleKinematic Self-Replicating Machines(LandesBioscience, 2004)First comprehensive review of field(c) 2004 Robert Freitas and Ralph MerkleFreitas is a technical consultant <strong>for</strong> this project1. The Concept of Self-ReplicatingMachines2. Classical Theory of Machine Replication3. Macroscale Kinematic MachineReplicators4. Microscale and Molecular KinematicMachine Replicators5. Issues in Kinematic Machine ReplicationEngineering6. Motivations <strong>for</strong> Molecular-Scale MachineReplicator Design
Related Work: Matt Moses• Strengths:• CAD to physicalimplementation• Large envelope UC• Weaknesses:• Strain bending under load• Manual controlMoses is a technical consultant <strong>for</strong> this project
Why Universal Constructors?RockAssembly LineRobotSRSUCEnvelope =nothingEnvelope =one thingEnvelope =many thingsEnvelope = everyconstituent partEnvelope =everything• UC is the ability to build anything• Uses “Genotype+Ribotype“= Phenotype”• Construction envelope includes itself• Atoms equivalent to bits• SRS only needs limited UC capability