エピソード

  • ep33 - Mathukumalli Vidyasagar: control synthesis, robotics, randomized algorithms, learning, compressed sensing, non-convex optimization

    ep33 - Mathukumalli Vidyasagar: control synthesis, robotics, randomized algorithms, learning, compressed sensing, non-convex optimization
    2025/06/16

    Outline

    00:00 - Intro

    00:42 - “Research should be fun”

    02:02 - Early steps in research

    09:00 - Book writing and meeting C. Desoer

    18:33 - Control synthesis via the factorization approach

    25:46 - The graph metric

    29:27 - Robotics and CAIR

    36:00 - Randomized algorithms

    40:41 - On learning

    44:05 - Neural networks

    48:40 - Tata, hidden Markov models, and large deviations theory

    55:48 - Picking problems and role of luck

    58:07 - Compressed sensing and non-convex optimization

    01:02:17 - Interplay between control and machine learning

    01:09:10 - Advice to future students

    01:13:29 - Future of control

    Links

    Sagar’s website: https://tinyurl.com/4hwruajs

    Hilbert: https://tinyurl.com/ykpdh929

    Feedback Systems: https://tinyurl.com/2k3jsdat

    How to Write Mathematics: https://tinyurl.com/35794bv9

    Nonlinear systems: https://tinyurl.com/2fdtnjcm

    C. Desoer: https://tinyurl.com/svhknren

    Control Systems Synthesis — A Factorization Approach: https://tinyurl.com/59wdc4sv

    Aryabhata: https://tinyurl.com/43x6hfhp

    A Brief History of the Graph Topology: https://tinyurl.com/49uftzdk

    Robot Dynamics and Control: https://tinyurl.com/5b4cmt7m

    CAIR: https://tinyurl.com/rajdtxax

    Randomized algorithms for robust controller synthesis using statistical learning theory: https://tinyurl.com/wanpyeuc

    R. Tempo: https://tinyurl.com/jkufdwar

    VC dimension: https://tinyurl.com/mvwk8afm

    Learning and Generalisation: https://tinyurl.com/2s3mzh8h

    Are Analog Neural Networks Better Than Binary Neural Networks? https://tinyurl.com/3fnk27xc

    Hidden Markov Processes: https://tinyurl.com/t5frrvfz

    An Introduction to Compressed Sensing: https://tinyurl.com/fc6a8eer

    Support the show

    Podcast info
    Podcast website: https://www.incontrolpodcast.com/
    Apple Podcasts: https://tinyurl.com/5n84j85j
    Spotify: https://tinyurl.com/4rwztj3c
    RSS: https://tinyurl.com/yc2fcv4y
    Youtube: https://tinyurl.com/bdbvhsj6
    Facebook: https://tinyurl.com/3z24yr43
    Twitter: https://twitter.com/IncontrolP
    Instagram: https://tinyurl.com/35cu4kr4

    Acknowledgments and sponsors
    This 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.
    続きを読む 一部表示
    1 時間 18 分
  • ep32 - Anders Rantzer: robust control, IQCs, nonlinear and hybrid systems, positivity and scalable control, adaptive control

    ep32 - Anders Rantzer: robust control, IQCs, nonlinear and hybrid systems, positivity and scalable control, adaptive control
    2025/05/16

    Outline
    00:00 - Intro and early steps in control
    06:42 - Journey to the US
    08:30 - Kharitonov's theorem and early influences
    12:10 - From Lund to KTH (Stockholm)
    15:01 - Ascona and collaboration with Megretski
    24:06 - The IMA year in Minnesota
    33:50 - Integral quadratic constraints
    46:10 - KYP lemma and meeting Yakubovich
    52:35 - Piecewise hybrid systems
    59:41 - Dual to Lyapunov theorem
    1:06:48 - Positivity and large scale systems
    1:15:25 - Adaptive and dual control
    1:25:24 - Future research directions

    Links
    Master's thesis: https://tinyurl.com/vnkuf44a
    Kharitonov's theorem: https://tinyurl.com/3w3a3z52
    IMA: https://cse.umn.edu/ima
    LCCC: http://www.lccc.lth.se/
    Real stability radius paper: https://tinyurl.com/yc79zt58
    IQC paper: https://tinyurl.com/mw2te4vw
    KYP lemma paper: https://tinyurl.com/2fv32yw6
    Piecewise quadratic Lyapunov functions for hybrid systems: https://tinyurl.com/38a5f432
    Piecewise linear quadratic optimal control: https://tinyurl.com/yhddbhpx
    Fast switches in relay feedback systems: https://tinyurl.com/23e95k9n
    Dual to Lyapunov's stability theorem: https://tinyurl.com/yc2uf652
    ECC’15 plenary: https://tinyurl.com/sy3k7vf7
    Scalable control of positive systems: https://tinyurl.com/3vcda447
    Tutorial on Positive Systems and Large Scale Control: https://tinyurl.com/3e4fkxz8
    KYP lemma for positive systems: https://tinyurl.com/3pdp3p7p
    CDC’25 plenary: https://tinyurl.com/497py5wh
    Data driven Riccati equation: https://tinyurl.com/3swjtvab

    Support the show

    Podcast info
    Podcast website: https://www.incontrolpodcast.com/
    Apple Podcasts: https://tinyurl.com/5n84j85j
    Spotify: https://tinyurl.com/4rwztj3c
    RSS: https://tinyurl.com/yc2fcv4y
    Youtube: https://tinyurl.com/bdbvhsj6
    Facebook: https://tinyurl.com/3z24yr43
    Twitter: https://twitter.com/IncontrolP
    Instagram: https://tinyurl.com/35cu4kr4

    Acknowledgments and sponsors
    This 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.
    続きを読む 一部表示
    1 時間 30 分
  • ep31 - Miroslav Krstić: nonlinear adaptive control, PDEs, delays, extremum seeking, safety, neural operators for control

    ep31 - Miroslav Krstić: nonlinear adaptive control, PDEs, delays, extremum seeking, safety, neural operators for control
    2025/04/15

    Outline
    00:00 - Intro
    01:07 - Early steps
    02:47 - Why control?
    05:20 - The move to the US
    07:40 - The first journal paper
    13:30 - What is backstepping?
    17:08 - Grad school
    25:10 - Stochastic stabilization
    29:53 - The interest in PDEs
    43:24 - Navier-Stokes equations
    52:12 - Hyperbolic PDEs and traffic models
    57:51 - Predictors for long delays
    1:08:14 - Extremum seeking
    1:27:14 - Safe control
    1:36:30 - Interplay between machine learning and control
    1:42:28 - Back to the roots: robust adaptive control
    1:50:50 - On service
    1:55:54 - Advice

    Links
    Miroslav’s site: https://flyingv.ucsd.edu/
    Tuning functions paper: https://tinyurl.com/yznv6r9r
    P. Kokotović: https://tinyurl.com/mwmbm9yh
    Separation and swapping: https://tinyurl.com/y4fre6t8
    Adaptive nonlinear stabilizers: https://tinyurl.com/4a9wmmvx
    KKK book: https://tinyurl.com/2kw2b4k6
    Stochastic nonlinear stabilization: https://tinyurl.com/4td3537a
    Follow-up with unknown covariance: https://tinyurl.com/4c4n7fd7
    Boundary state feedbacks for PIDEs: https://tinyurl.com/4e9y4tdr
    Boundary Control of PDEs: https://tinyurl.com/d8x38bmj
    Stabilization of Navier–Stokes systems: https://tinyurl.com/4a8cbjem
    Traffic congestion control: https://tinyurl.com/525jphs5
    Delay compensation: https://tinyurl.com/5yz6uj9p
    Nonlinear predictors for long delays: https://tinyurl.com/7wvce6vy
    Stability of extremum seeking: https://tinyurl.com/mr5cvzd3
    Nash equilibrium seeking: https://tinyurl.com/yeywrysn
    Inverse optimal safety filters: https://tinyurl.com/9dkrpvkk
    Neural operators for PDE control: https://tinyurl.com/5yynsp7v
    Bode lecture: https://tinyurl.com/mp92cs9u
    CSM article:

    Support the show

    Podcast info
    Podcast website: https://www.incontrolpodcast.com/
    Apple Podcasts: https://tinyurl.com/5n84j85j
    Spotify: https://tinyurl.com/4rwztj3c
    RSS: https://tinyurl.com/yc2fcv4y
    Youtube: https://tinyurl.com/bdbvhsj6
    Facebook: https://tinyurl.com/3z24yr43
    Twitter: https://twitter.com/IncontrolP
    Instagram: https://tinyurl.com/35cu4kr4

    Acknowledgments and sponsors
    This 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.
    続きを読む 一部表示
    2 時間 3 分
  • ep30 - Manfred Morari: A pioneer’s journey through robust, predictive and computational control

    ep30 - Manfred Morari: A pioneer’s journey through robust, predictive and computational control
    2025/02/17

    Outline
    00:00 - Intro
    03:26 - Development: ETH Zürich
    07:15 - Growth: Minnesota and Wisconsin
    36:16 - Productivity: Caltech
    53:28 - Change: ETH Zürich
    01:37:18 - Continuity: University of Pennsylvania
    01:45:36 - Outro

    Links
    Manfred’s website: https://tinyurl.com/mryp38w3
    Farewell lecture: https://tinyurl.com/2j5uxsp4
    Laying the foundations: an advisor’s perspective: https://tinyurl.com/ym5e4437
    Internal model control. A unifying review and some new results: https://tinyurl.com/vmm7a4tk
    Internal model control - PID controller design: https://tinyurl.com/rup3azjm
    Robust process control: https://tinyurl.com/8uhurkub
    Robust control of ill-conditioned plants: high-purity distillation: https://tinyurl.com/5xfty34c
    Computational complexity of \mu calculation: https://tinyurl.com/24yvvcc7
    Robust constrained model predictive control using linear matrix inequalities: https://tinyurl.com/3pdk55kk
    Control of systems integrating logic, dynamics, and constraints: https://tinyurl.com/489ekvsn
    The explicit linear quadratic regulator for constrained systems: https://tinyurl.com/mr2e3z83
    Model predictive control: https://tinyurl.com/2s4dafkd
    Parametric programming: https://tinyurl.com/2u9vj79y
    Embotech: https://tinyurl.com/3f99ks65
    Embedded online optimization for model predictive control at megahertz rates: https://tinyurl.com/59z3vnb9
    Donoho’s “Bridge from mathematical to digital and back”: https://tinyurl.com/3f9fk73w

    Support the show

    Podcast info
    Podcast website: https://www.incontrolpodcast.com/
    Apple Podcasts: https://tinyurl.com/5n84j85j
    Spotify: https://tinyurl.com/4rwztj3c
    RSS: https://tinyurl.com/yc2fcv4y
    Youtube: https://tinyurl.com/bdbvhsj6
    Facebook: https://tinyurl.com/3z24yr43
    Twitter: https://twitter.com/IncontrolP
    Instagram: https://tinyurl.com/35cu4kr4

    Acknowledgments and sponsors
    This 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.
    続きを読む 一部表示
    1 時間 46 分
  • ep29 - Richard Bellman: The Father of Dynamic Programming

    ep29 - Richard Bellman: The Father of Dynamic Programming
    2025/01/17

    Outline
    00:00 - Intro
    02:42 - Early life
    10:24 - WWII, Johns Hopkins, Belleville, Madison
    17:17 - Princeton
    20:30 - Los Alamos
    24:12 - Back to Princeton
    31:01 - RAND + Stanford
    42:10 - Dynamic programming
    46:06 - Project matterhorn
    51:00 - Back to RAND
    56:14 - Curse of dimensionality
    1:02:41 - Divorce and remarriage
    1:08:29 - Greenglass affair
    1:14:21 - USC
    1:18:55 - Center for the Study of Democratic Institutions
    1:22:55 - Last years
    1:26:08 - Outro

    Links
    R. Bellman: https://tinyurl.com/ydkmwexy
    C. M. Bellman: https://tinyurl.com/yfbafetj
    Dynamic programming: https://tinyurl.com/mtsuampr
    Curse of dimensionality: https://tinyurl.com/383yads8
    Hamilton-Jacobi-Bellman equation: https://tinyurl.com/muu4ckvs
    Bellman equation: https://tinyurl.com/y8bvzztm
    Grönwall-Bellman inequality: https://tinyurl.com/mrwdxzd8
    S. Lefschetz: https://tinyurl.com/2s39nnzv
    RAND corporation: https://tinyurl.com/4bfyhp9p
    Eye of the hurricane: https://tinyurl.com/bd6h84ka
    G. Szegő: https://tinyurl.com/mrthmmvm
    Project Matterhorn: https://tinyurl.com/5az9c3sh
    D. Greenglass: https://tinyurl.com/p2kfp7cz
    Center for the Study of Democratic Institutions: https://tinyurl.com/3vkupdty

    Support the show

    Podcast info
    Podcast website: https://www.incontrolpodcast.com/
    Apple Podcasts: https://tinyurl.com/5n84j85j
    Spotify: https://tinyurl.com/4rwztj3c
    RSS: https://tinyurl.com/yc2fcv4y
    Youtube: https://tinyurl.com/bdbvhsj6
    Facebook: https://tinyurl.com/3z24yr43
    Twitter: https://twitter.com/IncontrolP
    Instagram: https://tinyurl.com/35cu4kr4

    Acknowledgments and sponsors
    This 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.
    続きを読む 一部表示
    1 時間 30 分
  • ep28 - Karl Henrik Johansson: Semper in Motu - Hybrid Systems, Wireless and Event-Based Control, Mobility, Cybersecurity, and Societal Challenges

    ep28 - Karl Henrik Johansson: Semper in Motu - Hybrid Systems, Wireless and Event-Based Control, Mobility, Cybersecurity, and Societal Challenges
    2024/12/15

    Outline
    00:00 Intro
    01:19 Swedish control heritage and early steps
    05:21 PhD at Lund and relay feedback systems
    15:40 Berkeley years
    19:58 Hybrid systems
    26:46 Networked control systems
    30:32 Interaction with industry
    34:40 Wireless control systems
    40:33 Platooning in freight transport
    49:27 Future of mobility
    52:55 Event-based control
    58:58 Cybersecurity
    1:05:10 Societal-scale challenges
    1:12:33 Bode lecture reflections
    1:15:05 Digital futures
    1:23:38 Advice for future students

    Links
    Kalle’s website: https://tinyurl.com/yc34dv2x
    K. J. Åstrom: https://tinyurl.com/5n9admvk
    PhD Thesis: https://tinyurl.com/3868a92a
    Relay feedback: https://tinyurl.com/3w4w9dku
    Fast switches in relay feedback systems: https://tinyurl.com/5dcvn79f
    S. Sastry: https://tinyurl.com/3f5e978z
    Hybrid system: https://tinyurl.com/v39up6wk
    Dynamical properties of hybrid automata: https://tinyurl.com/yx73x6ba
    Wireless Network Design for Control Systems: https://tinyurl.com/mry5cnan
    Cyber–Physical Control of Road Freight Transport: https://tinyurl.com/yc4yhhwu
    S. Shladover: https://tinyurl.com/352cfwdm
    An introduction to event-triggered and self-triggered control: https://tinyurl.com/37p9w8cz
    Distributed Event-Triggered Control for Multi-Agent Systems: https://tinyurl.com/5ddnjsrz
    Cyber security analysis of state estimators in electric power systems: https://tinyurl.com/2wj74spj
    A secure control framework for resource-limited adversaries: https://tinyurl.com/37pnehuv
    Control for societal-scale challenges 2030: https://tinyurl.com/yfd3v296
    Digital futures: https://tinyurl.com/h3nyb2

    Support the show

    Podcast info
    Podcast website: https://www.incontrolpodcast.com/
    Apple Podcasts: https://tinyurl.com/5n84j85j
    Spotify: https://tinyurl.com/4rwztj3c
    RSS: https://tinyurl.com/yc2fcv4y
    Youtube: https://tinyurl.com/bdbvhsj6
    Facebook: https://tinyurl.com/3z24yr43
    Twitter: https://twitter.com/IncontrolP
    Instagram: https://tinyurl.com/35cu4kr4

    Acknowledgments and sponsors
    This 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.
    続きを読む 一部表示
    1 時間 29 分
  • ep27 - Munther Dahleh: L1 control, agile robotic maneuvering, abstractions, cascaded failures, markets, data and systems for societal problems

    ep27 - Munther Dahleh: L1 control, agile robotic maneuvering, abstractions, cascaded failures, markets, data and systems for societal problems
    2024/10/17

    Outline
    00:00 - Intro
    02:14 - Early steps in control
    05:36 - L1 control
    17:09 - Relay feedback systems
    24:32 - Agile robotic maneuvers
    29:31 - Abstractions and models
    41:59 - Causality
    49:35 - Markets and power grids
    1:00:20 - Role of network structure
    1:05:03 - Data, systems, and society
    1:16:17 - On writing
    1:22:57 - Cascaded failures in networks
    1:30:40 - Information design
    1:34:29 - Data markets
    1:42:27 - On economics and engineering
    1:48:18 - Systems thinking for societal problems
    1:58:24 - Advice to future students

    Links
    Website: https://dahleh.lids.mit.edu/
    Data Nation podcast: https://tinyurl.com/2fxwxw2d
    S. Bhattacharyya: https://tinyurl.com/4z93f6n7
    Wonham’s geometric control: https://tinyurl.com/mrxxkrv9
    B. Pearson: https://tinyurl.com/3vaz9yc9
    L1-optimal control: https://t.ly/NuXp8
    M. Vidyasagar: https://tinyurl.com/4z3zamh6
    Relay feedback systems: https://tinyurl.com/47tr23a4
    J. Tsitsilkis: https://tinyurl.com/yc7jmzm6
    Maneuver-based motion planning: https://tinyurl.com/bdduu2ea
    Nutonomy: https://tinyurl.com/426s7stv
    Feedback control with noisy channels: https://tinyurl.com/2yschy2s
    Bode’s integral: https://tinyurl.com/2s47rjmp
    Fundamental limits and molecular fluctuations: https://tinyurl.com/5n8k4maa
    Volatility of power grids: https://tinyurl.com/2vfpjfy7
    Bayesian learning in social networks: https://tinyurl.com/3mk9p486
    D. Acemoglu: https://tinyurl.com/3nphek9h
    A. Ozdaglar: https://tinyurl.com/yvdnddh9
    Cascade failures in networks: https://tinyurl.com/u2ckx94v
    A marketplace for data: https://tinyurl.com/esvbtprb


    Support the show

    Podcast info
    Podcast website: https://www.incontrolpodcast.com/
    Apple Podcasts: https://tinyurl.com/5n84j85j
    Spotify: https://tinyurl.com/4rwztj3c
    RSS: https://tinyurl.com/yc2fcv4y
    Youtube: https://tinyurl.com/bdbvhsj6
    Facebook: https://tinyurl.com/3z24yr43
    Twitter: https://twitter.com/IncontrolP
    Instagram: https://tinyurl.com/35cu4kr4

    Acknowledgments and sponsors
    This 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.
    続きを読む 一部表示
    2 時間 5 分
  • ep26 - Bart De Moor: subspace identification, linear & multilinear algebra, quantum information, ML

    ep26 - Bart De Moor: subspace identification, linear & multilinear algebra, quantum information, ML
    2024/08/21

    Outline
    00:00 - Intro
    00:49 - Early education and PhD in control
    10:35 - Non-negative linear algebra & modelling uncertainty principle
    22:50 - Influence of Willems & the Frisch scheme
    34:27 - Subspace identification & stochastic behavioural system theory
    38:39 - Stanford experience with Golub & Kailath
    53:30 - Return to Leuven: LS-SVMs, quantum information, and beyond
    1:07:48 - Multilinear algebra and tensor trains
    1:11:59 - Exact/approximate modelling and angles between systems
    1:25:34 - Back to the roots
    1:46:01 - On entrepreneurship: advice to future students
    1:56:32 - Outro

    Links
    Bart’s website: https://www.bartdemoor.be/
    Bart’s PhD Thesis: https://tinyurl.com/ryfu786h
    J. Willems: https://tinyurl.com/26zcrkc6
    From finite time series to linear system: https://tinyurl.com/4cvzpun6
    Subspace Identification For Linear Systems: https://tinyurl.com/5n8u228k
    Grassmann’s identity: https://tinyurl.com/37ey48da
    L. Ljung: https://tinyurl.com/2p9k74cm
    G. Golub: https://tinyurl.com/rteww7rb
    The restricted singular value decomposition: https://tinyurl.com/3pzhw8p4
    LS-SVMs: https://tinyurl.com/ykv7uvfv
    Four qubits can be entangled in nine different ways: https://tinyurl.com/y367kkcp
    High-order SVD: https://tinyurl.com/34u4p7h9
    Subspace angles between ARMA models: https://tinyurl.com/29cta3rm
    Exact and approximate modeling of linear systems: https://tinyurl.com/ykkbb9k4
    Back to the roots: https://tinyurl.com/ym6bvs38
    Resultant: https://tinyurl.com/32un7tbv
    Multidimensional system: https://tinyurl.com/jswpsr2s
    Macaulay resultant: https://tinyurl.com/y4skuevw
    Least squares realization of LTI models is an eigenvalue problem: https://tinyurl.com/3x9dmjnk


    Support the show

    Podcast info
    Podcast website: https://www.incontrolpodcast.com/
    Apple Podcasts: https://tinyurl.com/5n84j85j
    Spotify: https://tinyurl.com/4rwztj3c
    RSS: https://tinyurl.com/yc2fcv4y
    Youtube: https://tinyurl.com/bdbvhsj6
    Facebook: https://tinyurl.com/3z24yr43
    Twitter: https://twitter.com/IncontrolP
    Instagram: https://tinyurl.com/35cu4kr4

    Acknowledgments and sponsors
    This 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.
    続きを読む 一部表示
    1 時間 57 分