Podcast "inControl"

The first podcast on control theory and related topics, including feedback, decision making, artificial intelligence, robotics and much more. The podcast is supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» at ETH Zürich, which you can check at the following link: https://nccr-automation.ch/nccr-automation

Kategorien

Technik

Forschung

Webseite

Podcast "inControl"

Social Media

Panoptikum.social

Podcast-Episoden

ep21 Alberto Sangiovanni Vincentelli: Classics, EDA, Logic Synthesis, SPICE, Intel x386, Cadence, Synopsis, Platform-based design

Outline00:00 - Intro02:40 - Classic Studies06:51 - Early Steps in Berkeley between Optimization and Circuits Theory16:04 - Back to Italy... and back to Berkeley26:29 - The SPICE program and the interaction with IBM33:30 - Logic Synthesis, Simulated Annealing, Timberwolf40:57 - The Intel x386, YACR, and ESPRESSO47:01 - The birth of Cadence and Synopsis1:06:01 - Corsi e Ricorsi in EDA1:30:45 - The Shift to Automotive and Platform-Based Design1:39:06 - OutroLinksAlberto’s website: https://tinyurl.com/2w8t8fzaState-space approach in problem-solving optimization: https://tinyurl.com/bdcnmbhpEDA: https://tinyurl.com/3dur7ythSPICE: https://tinyurl.com/yxm56txyLogic Minimization Algorithms for VLSI Synthesis: https://tinyurl.com/49bnw3bySimulated annealing: https://tinyurl.com/mryz45pzTimberwolf: https://tinyurl.com/mvm28wnnIntel 386: https://tinyurl.com/4z7ubb55Cadence: https://tinyurl.com/hydjj92zSynopsis: https://tinyurl.com/463vk8mkDracula: https://tinyurl.com/kcum5v8tCorsi e ricorsi - the EDA story: https://tinyurl.com/mwar9p44G. Vico: https://tinyurl.com/cb4mp7nzFinite state machine: https://tinyurl.com/4ckszk2nSystem-level design: orthogonalization of concerns and platform-based design: https://tinyurl.com/v2t4njt4Quo Vadis, SLD? Reasoning About the Trends and Challenges of System Level Design: https://tinyurl.com/mpmmstunRemembering Richard: https://tinyurl.com/yc6rf7u3W. Shockley: https://tinyurl.com/w3chmuttSupport the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 15.03.2024
Dauer: 01:47:23

Weitere Informationen zur Episode "ep21 Alberto Sangiovanni Vincentelli: Classics, EDA, Logic Synthesis, SPICE, Intel x386, Cadence, Synopsis, Platform-based design"


ep20 - Aude Billard: From Robot learning to Dexterous Manipulation and Human-Robot Interaction

Outline00:00 - Intro00:58 - Early steps… in physics!05:17 - AI in Edinburgh08:00 - Connectionism10:37 - Robot learning19:02 - Imitation learning23:00 - On pursuing a PhD24:45 - Californian chronicles30:10 - Modularity33:20 - Challenges in robot learning39:44 - Dexterous manipulation43:52 - Dynamical systems47:37 - Combining ML and control51:25 - Human-robot interaction54:24 - Safety and compliance57:54 - Deadlines for papers1:06:38 - Advice to future students1:07:08 - OutroLinksAude’s lab: http://tinyurl.com/4nsx2kraRobot learning: http://tinyurl.com/2ty25t2rConnectionism: http://tinyurl.com/mrd76zfxI. Demiris: http://tinyurl.com/nhhn6ymuHebbian learning: http://tinyurl.com/35ey7um5DRAMA (paper): http://tinyurl.com/4fvbs867Imitation learning: http://tinyurl.com/2af3aew4Mirror neurons: http://tinyurl.com/4nd3sm75G. Rizzolatti: http://tinyurl.com/5b4xfab5Embodied cognition: http://tinyurl.com/5284kp8fA biologically inspired robotic model for learning by imitation: http://tinyurl.com/4j33nphvM. Airbib: http://tinyurl.com/3bu3y5dzRecent Advances in Robot Learning from Demonstration: http://tinyurl.com/487pm3xkSynthetic brain imaging (paper): http://tinyurl.com/38uerwk7Catching Objects in Flight (paper): http://tinyurl.com/38d8fk9mFeasibility not optimality (video): http://tinyurl.com/mwa8b7etLearning multiple-attractors (paper): http://tinyurl.com/2j3yxwm8On human compliance (paper): http://tinyurl.com/4vp2kmhwDeadlines for conferences: http://tinyurl.com/mpwntktySupport the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 16.02.2024
Dauer: 01:07:57

Weitere Informationen zur Episode "ep20 - Aude Billard: From Robot learning to Dexterous Manipulation and Human-Robot Interaction"


ep19 - Naomi Ehrich Leonard, Part II: Unveiling the Dynamics of Collective Decision-Making from Flocking Starlings to Desert Ants, Political Polarization, and the Creative Dance Between Arts and Control Theory

Outline00:00 - Intro01:33 - Starling Flock Formations07:17 - Harvesters ants in the desert16:54 - Decision making in the natural world and opinion dynamics 32:08 - A geometric look at political polarization36:50 - Control theory and dancing53:25 - CreativeX55:25 - On creativity57:47 - Advice to future studentsLinksAndrea Cavagna - http://tinyurl.com/uywn592jIrene Giardina - http://tinyurl.com/4jmk2h5n Starling Flock Networks Manage Uncertainty in Consensus at Low Cost: http://tinyurl.com/hc7wz5zpRegulation of harvester ant foraging: http://tinyurl.com/4s7samffFitz-Hugh-Nagumo model: http://tinyurl.com/yer32bhsBistability and Resurgent Epidemics in Reinfection Models: http://tinyurl.com/2p98fhpcSIR model: http://tinyurl.com/mrxdjvycMulti-agent system dynamics: Bifurcation and behavior of animal groups: http://tinyurl.com/ycx6ue2zFast and Flexible Multiagent Decision-Making: http://tinyurl.com/yd3azrjeA. Franci - http://tinyurl.com/5y5cn9yzA. Bizyaeva - http://tinyurl.com/3trp7c53M. Golubitsky - http://tinyurl.com/5ebkdtncI. Stewart - http://tinyurl.com/53nws9yzMulti-agent Decision-Making Dynamics Inspired by Honeybees: http://tinyurl.com/3jxmrkvdNonlinear Opinion Dynamics With Tunable Sensitivity: http://tinyurl.com/33mb23tbThe Nonlinear Feedback Dynamics of Asymmetric Political Polarization: http://tinyurl.com/rjdntrbpFlock logic: http://tinyurl.com/yyze25szThere may be others: http://tinyurl.com/fetjhecnSocial decision-making driven by artistic explore-exploit tension:  http://tinyurl.com/4r2uxebrRhythmbots: http://tinyurl.com/mr3k9f43CreativeX (Naomi's website): http://tinyurlSupport the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 15.01.2024
Dauer: 00:57:44

Weitere Informationen zur Episode "ep19 - Naomi Ehrich Leonard, Part II: Unveiling the Dynamics of Collective Decision-Making from Flocking Starlings to Desert Ants, Political Polarization, and the Creative Dance Between Arts and Control Theory"


Ep18 - Naomi Ehrich Leonard, Part I: Geometric Control on Lie Groups, Underwater Vehicles and Collective Motion, Coordination in Animal Groups, Decision Making in Honeybees and Bifurcation Theory

Outline00:00 - Intro01:05 - Dancing and control theory03:31 - Geometric control on Lie groups09:14 - Underwater vehicles and geometric mechanics18:45 - On the Hamiltonian framework21:25 - Underwater field experiments in Monte Rey Bay 36:27 - Collective motion and coordination in animal groups54:40 - Honeybees and bifurcation theory1:03:36 - OutroLinksNaomi’s website: http://tinyurl.com/j755aww5Naomi’s PhD Thesis: http://tinyurl.com/ywkvvy7kLie group: http://tinyurl.com/2p83jw9sAveraging: http://tinyurl.com/df9kmmcwStability of underwater vehicles: http://tinyurl.com/yxxytufxJ. Marsden: http://tinyurl.com/zvm8kkttA. Block: http://tinyurl.com/6wc39zkdCenter of buoyancy: http://tinyurl.com/mszncamhControlled Lagrangians: http://tinyurl.com/22usb52e - http://tinyurl.com/ymmntvr8Casimir function: http://tinyurl.com/yckc99mkMonterey Bay field experiments: http://tinyurl.com/yc24adct -  http://tinyurl.com/3sd7ee39 - http://tinyurl.com/ywryjwvr Collective motion: http://tinyurl.com/yuna5pam - http://tinyurl.com/pau74hmc - http://tinyurl.com/4p7zd5szSpatial patterns in coordinated groups: http://tinyurl.com/45y7hc9v-  http://tinyurl.com/5n7rm6vfKuramoto model: http://tinyurl.com/5eshfxhaDecision making in animal groups: http://tinyurl.com/3ybne8hn - http://tinyurl.com/283yts4y Value-Sensitive Decision-Making in honeybees: http://tinyurl.com/2uhcwyy6Bifurcation: http://tinyurl.com/tfr3ks7aSingularity theory: http://tinyurl.com/4kmxuv8zSupport the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 13.12.2023
Dauer: 01:07:16

Weitere Informationen zur Episode "Ep18 - Naomi Ehrich Leonard, Part I: Geometric Control on Lie Groups, Underwater Vehicles and Collective Motion, Coordination in Animal Groups, Decision Making in Honeybees and Bifurcation Theory"


ep17 - Tryphon Georgiou, Part II: Power spectra, optimal mass transport, Wasserstein geometry, turbulence, thermodynamics, quantum mechanics, adjustable one-ports

Outline00:00 - Intro00:47 - Mind the gap (metric)03:16 - Moment problems and Nevanlinna interpolation06:53 - “Everything relates to everything else”11:27 - Distances between power spectra 16:08 - Optimal mass transport and Schrödinger bridges32:25 - Sinkhorn iteration and Wasserstein geometry37:45 - Color of turbulence41:38 - Thermodynamics and energy harvesting from heat baths55:01 - Quantum mechanics57:55 - Adjustable one-ports1:07:20 - “Aha moments” and advice to future generations1:12:20 - OutroLinksTryphon’s website: https://georgiou.eng.uci.edu/Robustness analysis of nonlinear feedback systems: an input-output approach (paper): https://tinyurl.com/4785kxnyA topological approach to Nevanlinna–Pick interpolation (paper): https://tinyurl.com/4vudtz8cA generalized entropy criterion for Nevanlinna-Pick interpolation with degree constraint (paper): https://tinyurl.com/39enabucThe meaning of distances in spectral analysis, IEEE CDC plenary 2007 (slides): https://tinyurl.com/mrytp7j8Optimal Steering of a Linear Stochastic System to a Final Probability Distribution, Part I (paper): https://tinyurl.com/2nbm3sa6Y. Chen - https://tinyurl.com/37frfx67M. Pavon - https://tinyurl.com/yjhbawhbMittag Lefler - https://tinyurl.com/29cjum7jStochastic control liasons (paper): https://tinyurl.com/4s8y775bSinkhorn iteration: https://tinyurl.com/ym5catr2Color of turbulence (paper): https://tinyurl.com/5n77bepbHarvesting energy from a periodic heat bath (paper): https://tinyurl.com/2vadpu93Principles of lossless adjustable one-ports (paper): https://tinyurl.com/53v23yt4Inerter: https://tinyurl.com/ya2bkkhwSupport the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 11.12.2023
Dauer: 01:13:01

Weitere Informationen zur Episode "ep17 - Tryphon Georgiou, Part II: Power spectra, optimal mass transport, Wasserstein geometry, turbulence, thermodynamics, quantum mechanics, adjustable one-ports"


ep16 - Tryphon Georgiou, Part I: Marathons, Interpolation problems, Metrics, and the Arrow of Time

Outline00:00 - Intro01:50 - Running marathons 05:19 - The Center13:28 - On creativity15:24 - From algebraic system theory to moment problems43:39 - The gap metric58:33 - The longstanding friendship and collaboration with M. Smith01:11:30 - On causality and the arrow of time LinksTryphon’s website: https://georgiou.eng.uci.edu/People in control interview: https://tinyurl.com/4nw5s9p6R. Kalman: https://tinyurl.com/mux93t32A. Tannenbaum: https://tinyurl.com/2pws6rzdMoment problem - https://tinyurl.com/3u38xy9fNevanlinna–Pick interpolation - https://tinyurl.com/3nw56kjPh.D. Thesis: https://tinyurl.com/3c5ba8frOn the computation of the gap metric: https://tinyurl.com/tamnufmaUncertainty in Unstable Systems: The Gap Metric - https://tinyurl.com/4w7sn73nVidyasagar’s paper on the graph metric - https://tinyurl.com/5xn3rks6Optimal robustness in the gap metric - https://tinyurl.com/7axewjpyM. Smith - https://tinyurl.com/3ym2fbp9M. Vidyasagar - https://tinyurl.com/4fnwtjv7K. Glover - https://tinyurl.com/45zwpva9C. Foias - https://tinyurl.com/wxt378tjCommutant lifting theorem - https://tinyurl.com/bdfzxnf2D. Sarason - https://tinyurl.com/5n6n568fRobust Stability of Feedback Systems: A Geometric Approach Using the Gap Metric - https://tinyurl.com/bbv2hmy8Intrinsic difficulties in using the doubly-infinite time axis for input-output control theory - https://tinyurl.com/3cdbc9n2Erdős number - https://tinyurl.com/bdex5pf6Causal system - https://tinyurl.com/ythze2h7Feedback control and the arrow of time - https://tinyurl.com/2Support the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 14.11.2023
Dauer: 01:25:54

Weitere Informationen zur Episode "ep16 - Tryphon Georgiou, Part I: Marathons, Interpolation problems, Metrics, and the Arrow of Time"


ep15 - Davide Scaramuzza: Vision-Based Navigation, Agile Drone Racing, Perception-Aware Control, and Event Cameras

In this episode, we dive into the world of flying robots with Davide Scaramuzza, a leading expert in vision-based navigation, agile drone racing, perception-aware control, and the cutting-edge neuromorphic technology of event cameras. We explore the challenges of autonomous navigation in GPS-denied environments, the excitement of drone racing, the future of robotics, and the revolutionary potential of event-based cameras.Outline00:58 - Magic 02:58 - Visual SLAM and autonomous driving05:32 - Flying without a GPS11:01 - sFly project - Vision-based autonomous flight18:14 - Next steps22:30 - Drone racing and agile flying51:30 - Perception-aware control58:47 - On robustness1:02:46 - Risk-aware control and illumination1:07:52 - Event-based cameras1:15:37 - Agile flying  with event-based cameras1:19:28 - Event-based control and neuromorphic technology1:25:42 - Future of robotics1:30:55 - Advice to future students Links- Davide’s website: https://rpg.ifi.uzh.ch/people_scaramuzza.html- Copperfield at Niagara Falls: https://tinyurl.com/4wydc2s3- Ambitious card: https://tinyurl.com/5723kf8s- R. Siegwart: https://tinyurl.com/mr3sn472- sFly project: https://tinyurl.com/43hrffcx- DARPA challenge: https://tinyurl.com/5n7dnkmz- PTAM: https://tinyurl.com/epypbbmz- ROS: https://www.ros.org/- Acado: https://acado.github.io/- Drone racer - Nature paper: https://tinyurl.com/2rws2pjm- Drone racing - video: https://t.co/g9ckjV3O3N- Drone racing league: https://www.drl.io/ - Time-optimal MPCC: https://tinyurl.com/3udn5raf- Event-based vision: https://rpg.ifi.uzh.ch/research_dvs.html- T. Delbruck: https://tinyurl.com/4acymkxf- Event-based vision: a survey: https://tinyurl.com/2hwcmk9t- Event based vision and control paper: https:/Support the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 16.10.2023
Dauer: 01:33:47

Weitere Informationen zur Episode "ep15 - Davide Scaramuzza: Vision-Based Navigation, Agile Drone Racing, Perception-Aware Control, and Event Cameras"


ep14 - Cleve Moler: Numerical Analyst, First MATLAB Programmer, and MathWorks Co-Founder

In this episode, we chat with Cleve Moler, a pioneer in numerical mathematics,  creator of MATLAB and co-founder of MathWorks. We cover the birth of MATLAB, along with captivating stories about the origin of the iconic MathWorks logo, the enigmatic "why" command, the concept of "embarrassingly parallel computations," and the mysterious Pentium bug, among other. Outline00:00 - Intro  05:23 - Advice to students 05:45 - Caltech & J. Todd 07:07 - Stanford & G. Forsythe08:27 - The MathWorks logo  11:50 - ETH Zürich & Stiefel16:51 - Householder meetings 19:48 - LINPACK & EISPACK projects  26:10 - The birth of MATLAB 29:42 - Stanford course and the founding of Mathworks 38:40 - Embarrassingly parallel computing39:54 - The pentium bug 43:58 - SIAM and matrix exponentials47:19 - Future of mathematics51:36 - OutroLinksCleve’s corner - https://blogs.mathworks.com/cleve/Mathworks - https://mathworks.com/ History of Matlab - https://tinyurl.com/3dupkb7wDatatron computer - https://tinyurl.com/4kmcw95rJ. Todd - https://tinyurl.com/2s432wzcG. Forsythe - https://tinyurl.com/5583cfwxMathWorks logo - https://tinyurl.com/yc4th7sk E. Stiefel - https://tinyurl.com/ys4r2h96 J. Wilkinson - https://tinyurl.com/ye23bkdc LINPACK - https://tinyurl.com/39d7rwxk Computer solutions of linear algebraic systems - https://tinyurl.com/h9z7s342 Argonne Labs - https://www.anl.gov/ J. Dongarra - https://tinyurl.com/juzrw6y6 Embarrassingly parallel - https://tinyurl.com/yck38a4yPentium bug - https://tinyurl.com/4k7dt76p 19 dubious ways to compute the exponential of a matrix - https://tinyurl.com/yeyjy2bw Perron-Frobenius theorem - https://tinyurl.com/fa59dv32 O. Taussky - https://tinyurl.com/yckexuwsSupport the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 10.07.2023
Dauer: 00:52:37

Weitere Informationen zur Episode "ep14 - Cleve Moler: Numerical Analyst, First MATLAB Programmer, and MathWorks Co-Founder"


ep13 - John Doyle, Part II: Architectures, Universal laws, Layers, Levels, and Diversity-enabled Sweet Spots

In this episode, we chat again with John Doyle about the frontiers of control theory. Starting from the fascinating interplay between bacteria, physics, and the Internet, we explore the universal laws that govern complex systems. We discuss the inner workings of phenomena like earthquakes, wildfires, and sepsis, emphasizing the vital role of control theory in understanding intrinsic tradeoffs and fragilities. Finally, we discuss the essential components of a full theory of architectures,  including universal laws, layers, levels, and diversity-enabled sweet spots. Outline 00:00 - Intro 03:00 - Complex systems, physics, and the Internet 08:31 - On power laws 13:45 - SBML: Systems Biology Markup Language 18:51 - Layered architectures 21:38 - Earthquakes 26:17 - Wildfires 28:25 - Sepsis 37:18 - Essentials of a theory of architectures 54:10 - Universal laws, layers and levels 1:00:30 - Diversity enabled sweet spots  1:12:49 - Witsenhausen’s counterexample and SLS 1:21:25 - On the internal model principle 1:29:38 - Evolution vs intelligent design 1:33:37 - Fragility and societal implications 1:44:31 - OutroLinks Highly optimized tolerances and power laws paper: https://tinyurl.com/3yk2mycpRobust perfect adaptation in bacterial chemotaxis paper: https://tinyurl.com/3fn62a73SBML: https://sbml.org/Internet congestion control paper: https://tinyurl.com/4rjcd724The Structure of Scientific Revolutions: https://tinyurl.com/44n9y23uWildfires paper: https://tinyurl.com/2dvdh8apTurbulence paper: https://tinyurl.com/3sbsf8njSepsis paper: https://tinyurl.com/55wse56fDistributed LQG with delays paper: https://tinyurl.com/2abjdmb4Diversity-enabled sweet spots in layered architectures paper: https://tinyurl.com/vvaxvwb8Mountain biking game: https://tinyurl.com/46yh559rSystem-level synthesis paper: https://tinyurl.com/2ez64jevInternal feedback in biological control paper:  https://tinyurl.com/576zdfrxSupport the showPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, ETH studio and mirrorlake . Music was composed by A New Element.

Erschienen: 14.06.2023
Dauer: 01:46:38

Weitere Informationen zur Episode "ep13 - John Doyle, Part II: Architectures, Universal laws, Layers, Levels, and Diversity-enabled Sweet Spots"


ep12 - John Doyle, Part I: A Pioneer's Guide to Robust Control - The Past, Present, and Future

In this episode, we sit down with John Doyle, a living legend in the field of robust control, to delve into his incredible journey in control theory. We explore his past at MIT and Honeywell, his time at Berkeley, and his journey through the golden age of robustness. From his groundbreaking work on margins of systems, \mu synthesis, and the H_\infty problem, to his insights on System Level Synthesis (SLS) and modern control architectures, John shares his thoughts on the past, present, and future of robust control. Along the way, we listen to John's fascinating stories, including his astonishing sport records and his thrilling Panamanian adventure. Outline00:00 - Intro03:58 - Selected record-breaking athletics feats09:47 - The Panamanian adventure13:41 - Early steps in control: the MIT & Honeywell years 32:24 - The move to Berkeley and the golden age of robustness 46:06 -  To H_\infty and beyond 50:47 - DGKF: The solution of the H_\infty problem 1:02:40 - A glimpse of System Level Syntheis  (SLS) 1:07:27 - The challenge of our age: a theory of architecture design 1:12:34 - How to fix the theory-practice gap 1:15:05 - OutroLinksJohn’s website: https://doyle.caltech.edu/Main_PageSport records: https://tinyurl.com/4f7uapjt The Panamanian adventure: https://tinyurl.com/3zf4x5f7John’s master thesis: https://tinyurl.com/5c4bt5kkPaper - Guaranteed margins for LQG: https://tinyurl.com/3pjdvjmkPaper - Multivariable feedback design: ...  https://tinyurl.com/4uv8a6yzJohn’s PhD Thesis: https://tinyurl.com/27mew2kuPaper -  Feedback and optimal sensitivity: ... :  https://tinyurl.com/2p8a5vbhPaper - Performance and robustness analysis for structured uncertainty: https://tinyurl.com/mr78ajwxPaper - State-space solutions to standard H2 and H∞ control problems: https://tinyurl.com/4ru2ssc9Witsenhausen’s counterexample: https://tinyurl.com/3cavzz9yPodcast infoPodcast website: https://www.incontrolpodcast.com/Apple Podcasts: https://tinyurl.com/5n84j85jSpotify: https://tinyurl.com/4rwztj3cRSS: https://tinyurl.com/yc2fcv4yYoutube: https://tinyurl.com/bdbvhsj6Facebook: https://tinyurl.com/3z24yr43Twitter: https://twitter.com/IncontrolPInstagram: https://tinyurl.com/35cu4kr4Acknowledgments and sponsorsThis episode was supported by the National Centre of Competence in Research on «Dependable, ubiquitous automation» and the IFAC Activity fund. The podcast benefits from the help of an incredibly talented and passionate team. Special thanks to L. Seward, E. Cahard, F. Banis, F. Dörfler, J. Lygeros, as well as the ETH and mirrorlake studios. Music was composed by A New Element. Support the show

Erschienen: 15.05.2023
Dauer: 01:16:03

Weitere Informationen zur Episode "ep12 - John Doyle, Part I: A Pioneer's Guide to Robust Control - The Past, Present, and Future"


Podcast "inControl"
Merken
QR-Code