350 likes | 447 Views
Universal laws and architectures: Theory and lessons from brains, bugs, nets , grids, planes, docs, fire, bodies, fashion , earthquakes, turbulence , music, buildings, cities , art, running, throwing , S y n e s t h e s i a , spacecraft, statistical mechanics. John Doyle 道陽
E N D
Universal laws and architectures: Theory and lessons frombrains, bugs,nets, grids, planes, docs, fire, bodies, fashion, earthquakes, turbulence, music, buildings, cities, art, running, throwing, Synesthesia,spacecraft, statistical mechanics John Doyle 道陽 Jean-Lou Chameau Professor Controland Dynamical Systems, EE, & BioE 1 # Ca tech
Diverse applications Aerospace Robotics Biology Fluids Physics Medicine Internet Smartgrid Ecology Neuroscience Tuesday Wednesday Thursday Monday
Diverse applications Diverse mathematics ODE/PDE Automata Analysis Optimization Prob&Stats Operator Th Diff Geom AlgGeom CmpCplxty Aerospace Robotics Biology Fluids Physics Medicine Internet Smartgrid Ecology Neuroscience Details are not serious
Diverse mathematics (not just “applied”) Diverse applications (not just engineering) ODE/PDE Automata Analysis Optimization Prob&Stats Operator Th Diff Geom AlgGeom CmpCplxty Aerospace Robotics Biology Fluids Physics Medicine Internet Smartgrid Ecology Neuroscience CDS “Universal laws and architectures”
Diverse applications Domain experts have “core” interface to diverse math and tools Aerospace Robotics Biology Fluids Physics Medicine Internet Smartgrid Ecology Neuroscience CDS
Diverse mathematics Math (not just applied) has “core” interface to diverse application drivers ODE/PDE Automata Analysis Optimization Prob&Stats Operator Th Diff Geom AlgGeom CmpCplxty CDS
CDS as architecture • Architecture= “constraints that deconstrain” • Constraint = CDS core concepts and math • Deconstraint = connections between new applications and (also possibly new) math • How to explain this? • Laws: Universal constraints on achievable robust performance and efficiency • Architectures: Universal organizational strategies to flexibly achieve what is possible • Case studies: concrete and familiar examples to illustrate “universals”
Core protocols • From Architecture slides, understanding the “OS” is difficult and essential • But term “OS” seems to confuse • Replace with “core protocols”? • Highly conserved/constrained core (knot) • Highly diverse/deconstrained edges • What is “CDS” in this context?
Horizontal App Transfer Slow Apps OS HW Horizontal Hardware Transfer Fast Inflexible Flexible Special General
Tradeoffs Slow Apps OS HW Fast Inflexible Flexible Special General
Tradeoffs = “laws?” Slow Apps OS HW Fast Inflexible Flexible Special General
Layered architectures vision Prefrontal Learn Slow Motor Fast Sense HGT DNA repair Mutation DNA replication Transcription Translation Metabolism Signal Apps OS HW Dig. Lump. Distrib. OS HW Dig. Lump. Distrib. Digital Lump. Distrib. Lumped Distrib. Distrib. Evolve VOR Reflex Fast Inflexible Flexible Special General
Architecture = Constraints that deconstrain Horizontal App Transfer Deconstrained Constrained OS Deconstrained Horizontal Hardware Transfer
“Hourglass” Universal architectures Horizontal Transfer Deconstrained Constrained Deconstrained Horizontal Transfer
“Bowtie” Constrained Deconstrained Deconstrained Horizontal versus vertical is an arbitrary convention.
Diverse mathematics (not just “applied”) Diverse applications (not just engineering) ODE/PDE Automata Analysis Optimization Prob&Stats Operator Th Diff Geom AlgGeom CmpCplxty Aerospace Robotics Biology Fluids Physics Medicine Internet Smartgrid Ecology Neuroscience CDS “Universal laws and architectures”
“Pasteur’s quadrant?” (Donald Stokes) Typically poor choice of colors and font sizes
“Pasteur’s quadrant?” (Donald Stokes) High Use inspired basic research (Pasteur) Bohr Fundamental Understanding Edison Low High Low Practical use Relative emphasis
“Pasteur’s quadrant?” Redrawn Pasteur aimed for both Low Edison Fundamental Understanding Pasteur Bohr High High Low Relative emphasis Practical use
CDS Math Apps Fundamental Understanding Practical use
CDS’ quadrant? Specific Apps Scope CDS Math Deep, Universal Now Later Practical impact
Related but somewhat flipped SW Apps Slow OS HW Fast Inflexible Flexible Special General Specific Apps Scope CDS Math Deep, Universal Later Now Practical impact
Circa 1975 Specific Apps Modern control theory Scope Math Deep, Universal Now Later Practical impact
Complexity science Circa 1990 Specific Apps Modern control theory Robust control theory Scope Math Deep, Universal Now Later Practical impact
Nothing 2014 Network science Complexity science Apps Specific Scope CDS Math Deep, Universal Now Later Never Practical impact
Nothing Network science Complexity science Humans love this corner! Specific Scope Deep, Universal Now Later Never Practical impact
Network science Near future? Complexity science Specific Apps CMS CDS Math Deep, Universal Now Later Practical impact
Network science The dream? Complexity science Specific CMS Apps CDS Math Deep, Universal Now Later Practical impact
SW Apps Slow OS Universal laws Horizontal Transfer HW Deconstrained Fast Inflexible Flexible Special Constrained General Horizontal Transfer Deconstrained Universal architectures
Layered architectures vision Prefrontal Learn Slow Motor Fast Sense HGT DNA repair Mutation DNA replication Transcription Translation Metabolism Signal Apps OS HW Dig. Lump. Distrib. OS HW Dig. Lump. Distrib. Digital Lump. Distrib. Lumped Distrib. Distrib. Evolve VOR Reflex Fast Inflexible Flexible Special General
Diverse applications Morning “core” interface to minimal math and tools Aerospace Robotics Biology Fluids Physics Medicine Internet Smartgrid Ecology Neuroscience CDS Undergradmath Universal laws and architectures
Afternoon Diverse mathematics Cutting edge theory foundations Networks, Distributed ODE/PDE Automata Analysis Optimization Prob&Stats Operator Th Diff Geom AlgGeom CmpCplxty CDS
Network science The dream? Complexity science Specific CMS Apps CDS Math Deep, Universal Now Later Practical impact
Caveats • Many ideas are classic “robust control”, but much of the organization (the architecture) is fairly new • Motivating case studies also new • Many rough edges • Nobody is working on it quite like this • We want to change that