Pith. sign in

REVIEW 3 major objections 5 minor 49 references

Learning event-triggered controllers for linear parameter-varying systems from data

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A lifted-data rank condition turns noisy LPV trajectories into stabilizing event-triggered gain-scheduled controllers, with the same method extended to reference tracking.

desk verdict A load-bearing rank lemma for LPV data is unproved and its data-length bound contradicts the paper's own PE definition; the LMI synthesis is plausible but the foundation needs repair. read the letter →

arxiv 2506.08366 v1 pith:A3I3QPFD submitted 2025-06-10 eess.SY cs.SY

Renjie Ma , Su Zhang , Wenjie Liu , Zhijian Hu , Peng Shi This is my paper
classification eess.SYcs.SY
keywords data-drivencontrollinearparameter-varyingsystemsevent-triggeredpersistenceofexcitationPetersen'slemmafull-blockS-proceduregainschedulingreferencetracking
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper shows that a persistence-of-excitation condition on augmented LPV data makes it possible to certify closed-loop stability and synthesize stabilizing event-triggered gain-scheduled controllers directly from noisy input-state data, without identifying the system matrices. The key step is a new θ-persistence-of-excitation condition for LPV systems that guarantees the lifted data matrix has full row rank. With that rank guarantee, Petersen's lemma and the full-block S-procedure convert robust stability certificates into tractable semidefinite programs, solved only at the vertices of the scheduling set. The same construction, augmented with an integral compensator, handles output reference tracking. If the rank condition holds, feasible solutions yield both the controller gains and the event-triggering parameters that keep the closed loop practically exponentially input-to-state stable.

What carries the argument

The load-bearing object is the lifted data matrix Θ = Col(U, UP, X, XP), where UP and XP are the input and state sequences tensored with the scheduling signal, so that the unknown affine LPV matrices multiply Θ as in a linear regression. Lemma 1 asserts that if the augmented input sequence (input and its scheduling products) is θ-persistently exciting of order (1+ℓ)n+1, then Rank(Θ) = (1+ℓ)(n+m), provided the perturbation bound is small relative to the excitation level. That rank identity is what permits the data-based representations (11), (14), (37), and (60), which write next-state as the data matrix times a free parameter matrix. The stability certificates then rest on two classical tools: Petersen's lemma, which bounds the effect of the unknown bounded perturbation through the scalar ε, and the full-block S-procedure, which replaces the quadratically parameter-dependent matrix inequality by a finite set of LMIs at the vertices of the scheduling polytope. The event-triggering parameters Ψ1 and Ψ2 are extracted from a second SDP whose feasibility already presumes the stabilising gains from the first.

What would settle it

Take the paper's first simulation model and replace the randomly drawn scheduling sequence with a state-dependent one, for instance p_k = sign(x_{k,1}), while keeping the same θ-persistently exciting input; compute Rank(Θ). If the rank is ever below (1+ℓ)(n+m), Lemma 1 is violated and the data-based representations used in the theorems do not hold.

Watch

Extended reading notes

Core claim

The paper's central claim is that a perturbed discrete-time LPV system with affine dependence on a scheduling signal can be stabilized—and its output made to track a reference—using only collected input-state data and a known bound on the perturbation, with no identification of the system matrices. The claim is proven by establishing a θ-persistence-of-excitation condition for the augmented input (input stacked with its Kronecker products against the scheduling signal) that guarantees the lifted data matrix Θ has full row rank; this rank property makes valid the data-based closed-loop representations on which all later synthesis rests. From that representation, Petersen's lemma absorbs the unknown perturbation into a scalar certificate, and the full-block S-procedure converts the parameter-dependent matrix inequality into a finite family of linear matrix inequalities evaluated only at the vertices of the scheduling set. Feasible solutions of those LMIs produce the gain-scheduled state-feedback gains and, in a second program, the event-triggering parameters that make the closed loop practically exponentially input-to-state stable. The same construction, after augmenting the state with an integral of the tracking error, yields data-driven event-triggered reference tracking.

Load-bearing premise

The entire chain depends on Lemma 1's assertion that a persistently exciting augmented input forces the lifted data matrix Θ to full row rank, even though the scheduling signal p_k is not an independent input but is itself part of the state-dependent dynamics.

Editorial extensions

If this is right

  • A data set of length T ≥ n(1+ℓ)(1+m(1+ℓ))−1 with a θ-persistently exciting augmented input yields a full-row-rank matrix Θ, making the closed-loop data-based representation valid.
  • Feasible solutions of the vertex LMIs in Theorem 2 give gain-scheduled state-feedback gains Kd0 and K̄d that render the closed loop exponentially ISS for every perturbation inside the bound δ.
  • The same feasible solution yields event-triggering gains Ψ1, Ψ2; with the triggering rule ν_k^T Ψ1 ν_k ≥ x_k^T Ψ2 x_k + v, the closed loop is practically exponentially ISS and transmissions stop between triggers.
  • Augmenting the LPV system with an integral of the tracking error extends all certificates to output reference tracking, so the data-driven design covers trajectory following with the same vertex-LMI algorithm.
  • The two procedures require only the measured data, the known scheduling range, and the perturbation bound; no system-identification step or explicit model matrices are needed.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A dataset-level sanity check suggests itself: before solving any program, compute Rank(Θ) on the collected data; a rank shortfall invalidates every certificate regardless of how persistently exciting the input looks.
  • The paper's remark on measurement noise indicates that the Petersen-lemma step could be adapted to absorb sensor noise in the data as an enlarged perturbation, at the cost of a larger δ and tighter thresholds.
  • Choosing the event-triggering constant v adaptively from the data-based Lyapunov decrement rather than as a fixed user value would likely enlarge inter-event times while keeping the practical-ISS bound.
  • For LPV embeddings of genuinely nonlinear plants, the scheduling signal is usually a function of the state; verifying the θ-PE condition for such signals is exactly where the paper's independence assumption may be violated.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This manuscript proposes a direct data-driven, event-triggered controller synthesis for discrete-time LPV systems with bounded process disturbances. It introduces a theta-persistence-of-excitation condition intended to guarantee full row rank of a lifted data matrix, then derives stability certificates via Petersen's lemma and the full-block S-procedure, yielding SDPs with vertex-type constraints for gain-scheduled stabilization. A second set of results extends the method to reference tracking by adding an integral compensator and synthesizing an augmented-state LPV controller. The paper is illustrated by three simulation examples covering stabilization, scalar reference tracking, and planar trajectory tracking.

Significance. If the persistence-of-excitation/rank lemma were valid, this would be a useful contribution: it extends the recent data-driven event-triggered control results from LTI to LPV systems, it gives explicit SDP formulations, and it reports reproducible numerical experiments with code. The use of Petersen's lemma and full-block multipliers is appropriate in spirit, and the paper is well positioned relative to the recent LPV data-driven literature. The central caveat is that the rank lemma underpinning every data-based representation is not actually established, and the data-length statements in the examples are inconsistent with the stated PE hypothesis. These issues are load-bearing for the paper's main claims.

major comments (3)
  1. [Section II.B, Lemma 1] Lemma 1's proof applies [46, Theorem 3.1] to the lifted trajectory xbar_{k+1}=A_d zbar_k+B_d ubar_k, but zbar_{k+1}=Col(xbar_{k+1}, p_{k+1}⊗xbar_{k+1}) depends on the future scheduling p_{k+1}, so the lifted data are not generated by an LTI system and the robust fundamental lemma for LTI systems cannot be invoked. No condition on the scheduling sequence p_k is imposed and no LPV analog is proved, even though LPV fundamental-lemma results require additional conditions on the scheduling trajectory. Since Rank(Theta)=(1+ell)(n+m) is the prerequisite for equations (11), (14), (37), and (60), this gap undermines the central synthesis results. In addition, the sufficient condition involving theta, c_omega, delta, and rho depends on the unknown matrices A_d through c_omega, so it is not a data-verifiable PE certificate.
  2. [Lemma 1 and Examples 1 and 3] The stated data-length bound T ≥ n(1+ell)(1+m(1+ell))-1 is inconsistent with Definition 2. For U_k ∈ R^{m(1+ell)} and order L=(1+ell)n+1, the Hankel matrix H_L(U[0,T-1]) has m(1+ell)L rows and T-L+1 columns, so full row rank requires T ≥ (m(1+ell)+1)((1+ell)n+1)-1 = n(1+ell)(1+m(1+ell))+m(1+ell). Example 1 uses T=23 although this corrected bound gives T≥27, and Example 3 uses \hat T=29 although the corrected bound is 34 for nbar=3, m=2, ell=1. Thus the reported experiments do not satisfy the stated PE hypothesis, and simply checking Rank(Theta)=9 in Example 1 does not repair the missing full-rank property of the Hankel matrix.
  3. [Section III-D, Theorems 5-8] The reference-tracking results are not independently verifiable as written: Lemma 2 and Theorems 5-8 are each dismissed with 'the proof is along the same line' or 'we omit the details', and Theorem 5's LMI (62) contains a dimensionally inconsistent scalar (epsilon_1 in the (5,5) block instead of epsilon_3). Because tracking is one of the paper's principal claimed contributions and inherits the unproved rank lemma, the authors should provide the full augmented data construction for (57)-(60), the explicit PE condition on the augmented data, and the complete LMI derivations for Theorems 6-8.
minor comments (5)
  1. [Definition 1] The exponential ISS inequality (6) should contain a factor e^{-alpha_2 k} multiplying ||x_0||; as printed the decay term is independent of k and does not express exponential convergence.
  2. [Theorem 5, Eq. (62)] The bottom-right block of the LMI in (62) is epsilon_1 I_{\hat T}; it should be epsilon_3 I_{\hat T} to match the scalar introduced in the theorem.
  3. [Example 3] The matrices \hat A_{d0} and \hat B_{d0} are defined twice in Example 3; the second pair of definitions should be \hat A_{d1} and \hat B_{d1}.
  4. [Theorems 7 and 8] Theorem 7 states "\hat Psi_1, Psi_2" but both triggering matrices should be \hat Psi_1 and \hat Psi_2; additionally, the references to "Omega_{ii} defined in (40)/(72)" should point to (39)/(71), respectively.
  5. [Lemma 1 proof] The symbol X is used both for the perturbed data matrix in (4c) and for the nominal trajectory \bar{x} inside the proof, which makes the perturbation argument difficult to follow; distinct symbols should be used.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found; the synthesis results depend on standard external lemmas, and the main weakness is a technical derivation gap rather than circular reasoning.

full rationale

The paper's derivation chain is not circular. The central stability certificates are obtained from data-based representations via Schur complement, Petersen's lemma, and the full-block S-procedure; the theorems do not fit parameters to a target conclusion and then rename the fit as a prediction. Lemma 1 is the only load-bearing external step, and it invokes the robust fundamental lemma [46, Theorem 3.1], which is an independent external result by other authors; even if its application to lifted LPV data is questionable, that is a correctness or technical-support gap, not a self-referential reduction. The paper's self-citations (e.g., [22]) are contextual and not load-bearing. No uniqueness theorem from the authors' prior work is invoked to force the chosen controller structure, and no empirical pattern is merely renamed. The proofs of Lemma 2 and Theorems 5-8 are omitted with 'along the same line' statements, but omission of repetitive details is not circularity. The honest finding is that the manuscript is self-contained with respect to circularity, though the rank condition in Lemma 1 deserves scrutiny on external-applicability grounds.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The central claim rests on standard robust control lemmas and on a new PE condition whose proof is incomplete. All other quantities are user-specified bounds or decision variables from SDPs; no new physical entities are introduced.

free parameters (4)
  • perturbation bound delta = 0.1 (Example 1), 1.01 (Example 2), 2.51 (Example 3)
    Assumptions 1 and 2 require bounded energy noise; delta is a user-specified bound used in the LMIs. It is not fitted to data but is chosen by hand.
  • trigger threshold v = 0.01 (Example 1), 20 (Example 2), 500 (Example 3)
    Parameter in the triggering logic (38) and (70), chosen by hand to tune event frequency.
  • LMI scalars epsilon1, sigma, beta1, mu, epsilon2 = epsilon1=0.01, sigma=4, beta1=0.2, mu=40, epsilon2=0.001 in Example 1
    Tuning parameters in Theorems 1-4; chosen by hand to make the SDP feasible. Their values affect the certificates and trigger gains.
  • PE threshold theta = not specified numerically
    Lemma 1 assumes theta-persistence of excitation with theta > sqrt(...)c_omega delta / rho; the data length T is set to the lower bound, but theta itself is not computed in the examples.
assumptions (6)
  • standard math Petersen's lemma
    Used in proofs of Theorems 1 and 3 to handle bounded perturbation W with W W^T <= Delta Delta^T.
  • standard math Full-block S-procedure
    Used to reduce parameter-dependent LMIs to vertex conditions in Theorems 2 and 4.
  • domain assumption LPV system is controllable
    Lemma 1 and Lemma 2 assume controllability of the LPV system to invoke the fundamental lemma.
  • domain assumption Bounded perturbation W W^T <= Delta Delta^T with Delta = sqrt(T) delta I
    Assumptions 1 and 2 restrict the noise energy; all stability certificates depend on this bound.
  • ad hoc to paper Augmented-input persistence of excitation implies full row rank of Theta
    Lemma 1 states this but the proof transfers [46, Theorem 3.1] from LTI to LPV without justification. The paper treats it as the foundational prerequisite.
  • domain assumption Scheduling signal p_k lies in polytope P and is measurable online
    The LPV setup assumes p_k is in P; vertex relaxation in Theorems 2, 4, 6, and 8 requires a polytopic scheduling set.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning event-triggered controllers for linear parameter-varying systems from data." pith.science (2026). https://pith.science/paper/A3I3QPFD

@misc{pith2026250608366,
  author       = {Pith},
  title        = {Pith review of: Learning event-triggered controllers for linear parameter-varying systems from data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A3I3QPFD}},
  note         = {Machine review of arXiv:2506.08366}
}
abstract

Nonlinear dynamical behaviours in engineering applications can be approximated by linear-parameter varying (LPV) representations, but obtaining precise model knowledge to develop a control algorithm is difficult in practice. In this paper, we develop the data-driven control strategies for event-triggered LPV systems with stability verifications. First, we provide the theoretical analysis of ${\theta}$-persistence of excitation for LPV systems, which leads to the feasible data-based representations. Then, in terms of the available perturbed data, we derive the stability certificates for event-triggered LPV systems with the aid of Petersen's lemma in the sense of robust control, resulting in the computationally tractable semidefinite programmings, the feasible solutions of which yields the optimal gain schedulings. Besides, we generalize the data-driven eventtriggered LPV control methods to the scenario of reference trajectory tracking, and discuss the robust tracking stability accordingly. Finally, we verify the effectiveness of our theoretical derivations by numerical simulations.

Figures

Figures reproduced from arXiv: 2506.08366 by the authors.

Figure 1
Figure 1. The design schematic of event-triggered LPV control from data. In what follows, we aim to perform the control synthesis of perturbed LPV systems with closed-loop stability verifications, leading to the data-driven control strategies applicable to both robust stabilization and reference tracking. On the premise of proposed θ-persistence of excitation prerequisite of perturbed LPV systems in Lemma 1, the problems to b… view at source ↗
Figure 2
Figure 2. The state responses and inter-event intervals of event-triggered LPV systems (37) actuated by data-driven controller (34). In what follows, we aim to perform two examples to validate the effectiveness of Procedure 2. Example 2 corresponds to the scenarios of one-dimension output tracking, while Example 3 corresponds to the plane output tracking, which are potentially utilized for robotic motion control [PITH_FULL_I… view at source ↗
Figure 3
Figure 3. The robust tracking responses and inter-event intervals of event￾triggered LPV systems (53) actuated by data-driven controller (66)-Case 1: sinusoidal reference signal. 0 1 2 3 4 5 6 Time (Sec.) -2 -1 0 1 2 0 0.05 -1 -0.5 0 0.2 0.4 0.6 0.8 1 0 1 2 3 4 5 6 Time (Sec.) [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: The robust tracking responses and inter-event intervals of event￾triggered LPV systems (53) actuated by data-driven controller (66)-Case 2: square-wave signal reference signal. leads to a feasible solution P =   0.0486 −0.0022 −0.0065 −0.0022 0.0248 0.0000 −0.0065 0.…
Figure 5
Figure 5. Figure 5: The robust tracking responses and inter-event intervals of event-triggered LPV systems (53) actuated by data-driven controller (66)-Case 1: circle reference and Case 2: shape-8 reference. [29] T. Dai and M. Sznaier, ”A semi-algebraic optimization approach to data-drive…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

49 extracted references · 49 canonical work pages

  1. [1]

    Mohammadpour and C

    J. Mohammadpour and C. W. Scherer, Control of Linear Parameter V arying Systems with Applications. Springer Science & Business Media, 2012

    work page 2012

  2. [2]

    Martin, T

    T. Martin, T. B. Sch ¨on, and F. Allg ¨ower, ”Guarantees for data-driven control of nonlinear systems using semidefinite programmin g: A sur- vey”, Annual Reviews in Control , vol. 56, num. 100911, 2023

    work page 2023

  3. [3]

    X. Ping, J. Y ao, B. Ding, and Z. Li, ”Tube-based output feedba ck robust MPC for LPV systems with scaled terminal constraint s ets”, IEEE Transactions on Cybernetics , vol. 52, no. 8, pp. 7563-7576, 2022

    work page 2022

  4. [4]

    A. P . Pandey and M. C. de Oliveira, ”A new discrete-time stabi lizability condition for linear parameter-varying systems”, Automatica, vol. 79, pp. 214-217, 2017

    work page 2017

  5. [5]

    A. N. V argas, C. M. Agulhari, R. C. L. F. Oliveira, and V . M. Pre ciado, ”Robust stability analysis of linear parameter-varying sy stems with Markov jumps”, IEEE Transactions on Automatic Control , vol. 67, no. 11, pp. 6234-6239, 2022. 1The codes are available by https://github.com/Renjie-Ma/ Learning-Event- Triggered-Controller-LPV -Systems-From-Data.git

    work page 2022

  6. [6]

    P . B. Cox, S. Weiland, and R. T ´oth, ”Affine parameter-dependent Lyapunov functions for LPV systems with affine dependence”, IEEE Transactions on Automatic Control, vol. 63, no. 11, pp. 3865-3872, 2018

    work page 2018

  7. [7]

    J. A. Gallegos and K. A. Barbosa, ”Set-theoretical stabilit y analysis of LPV systems via Minkowski-Lyapunov functions”, IEEE Transactions on Automatic Control , DOI: 10.1109/TAC.2024.3441325

    work page arXiv 2024

  8. [8]

    S. K. Mulagaleti, M. Mejari, and A. Bemporad, ”Parameter-de pendent robust control invariant sets for LPV systems with bounded p arameter- variation rate”, IEEE Transactions on Automatic Control , DOI: 10.1109/TAC.2024.3454528

    work page arXiv 2024

Show all 49 references
  1. [9]

    Cheng, M

    J. Cheng, M. Wu, F. Wu, C. Lu, X. Chen, and W. Cao, ”Modeling and control of drill-string system with stick-slip vibrati ons using LPV technique”, IEEE Transactions on Control Systems Technology , vol. 29, no. 2, pp. 718-730, 2021

    2021

  2. [10]

    Polcz, B

    P . Polcz, B. Kulcsar, T. Peni, and G. Szederkenyi, ”Passivit y analysis of rational LPV systems using Finsler’s lemma”, in IEEE 58th Conference on Decision and Control , pp. 3793-3798, 2019

    2019

  3. [11]

    C. E. de Souza, K. A. Barbosa, and A. T. Neto, ”Robust H∞ filtering for discrete-time linear systems with uncertain time-vary ing parameters”, IEEE Transactions on Signal Processing , vol. 54, no. 6, pp. 2110-2118, 2006

    2006

  4. [12]

    G. Cai, T. Wu, M. Hao, H. Liu, and B. Zhou, ”Dynamic event-trig gered gain-scheduled H∞ control for a polytopic LPV model of morphing aircraft”, IEEE Transactions on Aerospace and Electronic Systems , vol. 61, no. 1, pp. 93-106, 2025

    2025

  5. [13]

    T. J. Meijer, V . Dolk, and W. P . M. H. Heemels, ”Certificates of nonexistence for analyzing stability, stabilizability an d detectability of LPV systems”, Automatica, vol. 170, num. 111841, 2024

    2024

  6. [14]

    J. C. Willems, P . Rapisarda, I. Markovsky, and B. L. M. De Moor , ”A note on persistency of excitation”, Systems & Control Letters , vol. 54, pp. 325-329, 2005

    2005

  7. [15]

    K. He, S. Shi, T. van den Boom, and B. De Schutter, ”From learni ng to safety: A direct data-driven framework for constrained con trol”, arXiv Preprint, arXiv: 2505.15515v1

  8. [16]

    W. Liu, G. Wang, J. Sun, F. Bullo, and J. Chen, ”Learning robus t data-based LQG controllers from noisy data”, IEEE Transactions on Automatic Control, vol. 69, no. 12, pp. 8526-8538, 2024

    2024

  9. [17]

    De Persis and P

    C. De Persis and P . Tesi, ”Formulas for data-driven control: Stabilization, optimality, and robustness”, IEEE Transactions on Automatic Control , vol. 65, no. 3, pp. 909-924, 2020

    2020

  10. [18]

    D ¨orfler, P

    F. D ¨orfler, P . Tesi, and C. De Persis, ”On the certainty-equivale nce approach to direct data-driven LQR design”, IEEE Transactions on Automatic Control, vol. 68, no. 12, pp. 7989-7996, 2023

    2023

  11. [19]

    Berberich, J

    J. Berberich, J. K ¨ohler, M. A. M ¨uller, and F. Allg ¨ower, ”Data-driven model predictive control with stability and robustness gua rantees”, IEEE Transactions on Automatic Control, vol. 66, no. 4, pp. 1702-1717, 2021

    2021

  12. [20]

    Z. Ren, H. Liu, G. Wen, and J. L ¨u, ”Event-triggered data-driven security formation control for quadrotors under denial-of -service attacks and communication faults”, IEEE Transactions on Cybernetics , DOI: 10.1109/TCYB.2024.3467178

  13. [21]

    J. Wang, J. Sun, J. Y ang, and S. Li, ”Periodic event-triggere d model predictive control for networked nonlinear uncertain syst ems with dis- turbances”, IEEE Transactions on Cybernetics, vol. 54, no. 12, pp. 7501- 7513, 2024

    2024

  14. [22]

    R. Ma, P . Shi, and L. Wu, ”Sparse false injection attacks reco nstruction via descriptor sliding mode observers”, IEEE Transactions on Automatic Control, vol. 66, no. 11, pp. 5369-5376, 2021

    2021

  15. [23]

    T. An, B. Dong, H. Y an, L. Liu, and B. Ma, ”Dynamic event- triggered strategy-based optimal control of modular robot manipulator: A multiplayer nonzero-sum game perspective”, IEEE Transactions on Cybernetics, vol. 54, no. 12, pp. 7514-7526, 2024

    2024

  16. [24]

    Rotulo, C

    M. Rotulo, C. De Persis, and P . Tesi, ”Online learning of data -driven controllers for unknown switched linear systems”, Automatica, vol. 145, num. 110519, 2022

    2022

  17. [25]

    J.Berberich, C. W. Scherer, and F. Allg ¨ower, ”Combining prior knowl- edge and data for robust controller design”, IEEE Transactions on Automatic Control, vol. 68, no. 8, pp. 4618-4633, 2023

    2023

  18. [26]

    H. J. van Waarde, M. K. Camlibel, and M. Mesbahi, ”From noisy d ata to feedback controllers: Nonconservative design via a matr ix S-lemma”, IEEE Transactions on Automatic Control , vol. 67, no. 1, pp. 162-175, 2022

    2022

  19. [27]

    Bisoffi, C

    A. Bisoffi, C. De Persis, and P . Tesi, ”Data-driven control vi a Petersen’s lemma”, Automatica, vol. 145, num. 110537, 2022

    2022

  20. [28]

    De Persis, M

    C. De Persis, M. Rotulo, and P . Tesi, ”Learning controllers f rom data via approximate nonlinearity cancellation”, IEEE Transactions on Automatic Control, vol. 68, no. 10, pp. 6082-6097, 2023. AUTHOR et al.: TITLE 13 -4 -3 -2 -1 0 1 2 3 4 -3 -2 -1 0 1 2 3 (a) The circle traje...

    2023

  21. [30]

    A. Koch, J. Berberich, and F. Allg ¨ower, ”Provably robust verification of dissipativity properties from data”, IEEE Transactions on Automatic Control, vol. 67, no. 8, pp. 4248-4255, 2022

    2022

  22. [31]

    Bisoffi, C

    A. Bisoffi, C. De Persis, and P . Tesi, ”Learning controllers f or perfor- mance through LMI regions”, IEEE Transactions on Automatic Control , vol. 68, no. 7, pp. 4351-4358, 2023

    2023

  23. [32]

    L. Li, C. De Persis, P . Tesi, and N. Monshizadeh, ”Data-based transfer stabilization in linear systems”, IEEE Transactions on Automatic Con- trol, vol. 69, no. 3, pp. 1866-1873, 2024

    2024

  24. [33]

    Str¨asser, M

    R. Str¨asser, M. Schaller, K. Worthmann, J. Berberich, and F. Allg ¨ower, ”Koopman-based feedback design with stability guarantees ”, IEEE Transactions on Automatic Control , vol. 70, no. 1, pp. 355-370, 2025

    2025

  25. [34]

    A. Koch, J. Berberich, J. K ¨ohler, and F. Allg¨ower, ”Determining optimal input-output properties: A data-driven approach”, Automatica, vol. 134, num. 109906, 2021

    2021

  26. [35]

    Bianchi, S

    M. Bianchi, S. Grammatico, and J. Cort ´es, ”Data-driven stabilization of switched and constrained linear systems”, Automatica, vol. 171, num. 111974, 2025

    2025

  27. [36]

    Disar `o and M

    G. Disar `o and M. E. V alcher, ”On the equivalence of model-based and data-driven approaches to the design of unknown-input obse rvers”, IEEE Transactions on Automatic Control, vol. 70, no. 3, pp. 2074-2081, 2025

    2025

  28. [37]

    De Persis, R

    C. De Persis, R. Postoyan, and P . Tesi, ”Event-triggered con trol from data”, IEEE Transactions on Automatic Control, vol. 69, no. 6, pp. 3780- 3795, 2024

    2024

  29. [38]

    H. Zhou, Y . Zuo, and S. Tong, ”Fuzzy adaptive event-triggere d con- sensus control for nonlinear multiagent systems under join tly connected switching networks”, IEEE Transactions on Cybernetics , vol. 54, no. 12, pp. 7163-7172, 2024

    2024

  30. [39]

    C. Liu, Z. Chu, Z. Duan, H. Zhang, and Z. Ma, ”Decentralized ev ent- triggered tracking control for unmanned interconnected sy stems via particle swarm optimization-based adaptive dynamic progr amming”, IEEE Transactions on Cybernetics, vol. 54, no. 11, pp. 6895-6909, 2024

    2024

  31. [40]

    Wildhagen, J

    S. Wildhagen, J. Berberich, M. Hertneck, and F. Allg ¨ower, ”Data-driven analysis and controller design for discrete-time systems u nder aperiodic sampling”, IEEE Transactions on Automatic Control , vol. 68, no. 6, pp. 3210-3225, 2023

    2023

  32. [41]

    V erhoek, J

    C. V erhoek, J. Berberich, S. Haesaert, F. Allg ¨ower, and R. T ´oth, ”Data- driven dissipativity analysis of linear parameter-varyin g systems”, IEEE Transactions on Automatic Control, vol. 69, no. 12, pp. 8603-8616, 2024

    2024

  33. [42]

    V erhoek, R

    C. V erhoek, R. T´oth, and H. S. Abbas, ”Direct data-driven state-feedback control of linear parameter-varying systems”, arXiv Preprint , arXiv: 2211.17182v4

    arXiv

  34. [43]

    V erhoek, I

    C. V erhoek, I. Markovsky, S. Haesaert, and R. T ´oth, ”The behavioral approach for LPV data-driven representations”, arXiv Preprint , arXiv: 2412.18543v1

  35. [44]

    H. S. Abbas, R. T ´oth, N. Meskin, J. Mohanmadpour, and J. Hanema, ”A robust MPC for input-output LPV models”, IEEE Transactions on Automatic Control, vol. 61, no. 12, pp. 4183-4188, 2016

    2016

  36. [45]

    Golabi, N

    A. Golabi, N. Meskin, R. T ´oth, J. Mohammadpour, and T. Donkers, ”Event-triggered control for discrete-time linear parame ter-varying sys- tems”, in 2016 American Control Conference , pp. 3680-3685, 2016

    2016

  37. [46]

    Clouson, H

    J. Clouson, H. van Waarde, and F. D ¨orfler, ”Robust fundamental lemma for data-driven control”, arXiv Preprint , arXiv: 2205.06636

    arXiv

  38. [47]

    Digge and R

    V . Digge and R. Pasumarthy, ”Data-driven event-triggered c ontrol for discrete-time LTI systems”, in 2022 European Control Conference , pp. 1355-1360,

    2022

  39. [48]

    C. W. Scherer, ”LPV control and full block multipliers”, Automatica, vol. 37, pp. 361-375, 2001

    2001

  40. [49]

    L ¨ofberg, ”Y ALMIP: A toolbox for modeling and optimization in MA TLAB”, in 2004 IEEE International Confernece on Robotics and Automation, pp

    J. L ¨ofberg, ”Y ALMIP: A toolbox for modeling and optimization in MA TLAB”, in 2004 IEEE International Confernece on Robotics and Automation, pp. 284-289, 2004

    2004

  41. [50]

    ApS, Mosek Optimization Toolbox for Matlab, User’s Guide and Reference Manual, V ersion 4, [Online] https://www.mosek.com/, 2019

    M. ApS, Mosek Optimization Toolbox for Matlab, User’s Guide and Reference Manual, V ersion 4, [Online] https://www.mosek.com/, 2019

    2019

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.