Pith. sign in

REVIEW 6 major objections 4 minor 32 references

Dimension-varying control systems can be modelled on a single metric space Ω, built by identifying vectors that differ only by repeating coordinates, so trajectories may cross dimensions continuously.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

A quotient space Ω with a repetition-based metric turns dimension-varying control systems into ordinary switched systems on one common state space.

T0 review reviewed 2026-08-03 challenge →

load-bearing objection Repackages Cheng's cross-dimensional metric into control problems; coherent topology but the repetition equivalence is unvalidated and the control theorems are asserted rather than proved. the 6 major comments →

arxiv 2601.14839 v2 pith:2INPFSXG submitted 2026-01-21 math.OC

On Dimension-Varying Control Systems: A Universal State Space Approach

classification math.OC MSC 93C3093B0593B0793D1554B15
keywords dimension-varying control systemscross-dimensional Euclidean spaceuniversal state spacedimension-keeping semi-tensor productLipschitz switchingcontrollabilityobservabilitydisturbance decoupling
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The reading

Dimension-varying systems—networks whose agents join or leave, power grids that shed or reconnect generators, spacecraft that dock and undock—have no common geometry: the state lies in R^n at one instant and in R^m at the next. This paper claims that a single metric space, Ω, resolves the problem: declare x∈R^m and y∈R^n equivalent when repeating each coordinate an integer number of times makes them identical, so (a,b) and (a,a,b,b) are the same point. The quotient space Ω carries a metric, each R^n embeds homeomorphically as a subspace Ω_n, and trajectories can be continuous across dimension changes. In Ω the paper formulates controllability, observability, stabilization, and disturbance decoupling, turning a dimension-varying system into an equivalent switching system with impulses. If the identification matches physical reality, standard switching-system results apply uniformly to variable-dimension dynamics.

Core claim

The paper's central claim is that R^∞ = ∪ R^n, equipped with the cross-dimensional inner product ⟨x,y⟩_V = (1/t)⟨x⊗1_{t/m}, y⊗1_{t/n}⟩ and the resulting distance d_V, becomes the universal state space for dimension-varying systems after quotienting by zero-distance equivalence. The key structural fact is that each Ω_n = Ω∩R^n is homeomorphic to R^n with the Euclidean topology, so embedding a mode into Ω changes nothing locally, while globally Ω is path-connected through the intersections Ω_{m∧n}. On this space the paper defines Lipschitz switchings, common Lyapunov functions for stability, controllability along lattice chains, mode-wise observability, stabilization with or without jumps, dis

What carries the argument

The central object is the quotient space Ω = R^∞/↔, where x↔y iff x⊗1_p = y⊗1_q for some repetition factors p,q, equivalently d_V(x,y)=0. The equivalence identifies a vector with any blockwise replication of itself, so dimensions become commensurable through least common multiples. Two pieces of machinery carry the argument: the cross-dimensional projection Π^m_n = (m/t)(I_m⊗1^T_{t/m})(I_n⊗1_{t/n}), which sends a vector in R^n to its nearest point in R^m under d_V and doubles as the bridge matrix of the dimension-keeping semi-tensor product; and the lattice structure on {Ω_n} ordered by divisibility, whose sup and inf are lcm and gcd. The projection makes jumps well-defined, such as nearest-

Load-bearing premise

The load-bearing premise is that repeating a state's coordinates—turning (a,b) into (a,a,b,b)—creates the same physical state, so systems that distinguish a 2-component configuration from its 4-component replication are being modeled in an artificial space.

What would settle it

Take a concrete mechanical system with one mode in R^2 and another in R^4, with the transition energy or output depending on the actual number of coordinates—for instance, two coupled oscillators versus four identical oscillators at positions (a,a,b,b). Measure the switching cost between x=(a,b) and y=(a,a,b,b): d_V(x,y)=0, so Ω treats them as the same state, while any nonzero energy difference, distinct output, or different future trajectory would violate the equivalence and refute the metric's physical claims. Even one clean demonstration of such a distinction in a domain the paper claims to

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • A dimension-varying control system can be embedded in Ω with no modification of the local dynamics: each mode's R^{n_i} is homeomorphic to Ω_{n_i}, so local stability and controllability statements are preserved.
  • Switching between modes can be treated as a jump in a single metric space; if the jump is Lipschitz, a common Lyapunov function plus sufficiently large minimum dwell time yields stability.
  • Controllability can be established mode-by-mode along a lattice chain: partial controllability on the orthogonal complement of the intersection Ω_{n_i}∩Ω_{n_j} and complete controllability on the final mode suffice.
  • Disturbance decoupling reduces to checking each lifted mode after projecting the disturbance into that mode's dimension; components orthogonal to the mode are harmless.
  • The same construction converts dimension-varying systems into classical switching systems with Dirac impulses, so standard switched-control techniques apply.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If d_V is taken seriously, the framework predicts a specific testable symmetry: any physical variable replicated by an integer factor should be physically indistinguishable from the original. For mechanical systems this is rarely true—two identical masses at (a,a,b,b) generally carry different energy and dynamics than one mass at (a,b)—so the metric's domain of validity is likely signal-like or co
  • The lattice structure suggests a natural extension to asynchronous or partially coupled dimension changes: subsystems could move along different sublattices, with consistency enforced only at joins, and the paper's nearest-jump and projection machinery would carry over almost unchanged.
  • A quantitative extension would replace the hard equivalence d_V=0 with a tolerance: a controlled system stabilizes to within ε in Ω rather than exactly, making 'practically controllable' discontinuous switches rigorous and testable.
  • The information-theoretic reading at the end—repeated coordinates carry the same entropy—points toward applications in decentralized estimation or compression, where the state is a distribution or message and replication genuinely carries no new information.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

6 major / 4 minor

Summary. This paper proposes a quotient space Ω=R^∞/↔ with a cross-dimensional metric d_V as a universal state space for dimension-varying control systems, where x↔y iff x⊗1_p=y⊗1_q. It proves each R^n embeds homeomorphically as Ω_n, develops a lattice of subspaces, defines Lipschitz switching and stability, and claims controllability, observability, stabilizability, disturbance decoupling, and hierarchical aggregation results in Ω. The central advertised claim is that fixed-dimension control problems can be transplanted to the dimension-varying setting without external embeddings.

Significance. The construction is attractive and partly rigorous: d_V restricted to R^n equals (1/√n)‖·‖_2, so Theorem 3.4 gives a clean homeomorphism; the projection formula (7) and operator norm formula (24) are explicit and checkable; the triangle-geometry example (Example 3.6) is a nice sanity check. The paper also honestly flags the main modeling caveat in the conclusion. If the control theorems were proved in detail, the framework would be a useful common language for switched/dimension-varying systems. At present, however, the universal claim rests on an unvalidated equivalence relation and on several proof sketches, so the significance is prospective rather than established.

major comments (6)
  1. [Sec. II (Defs. 2.1–2.3), Sec. VIII] The identification x↔y iff x⊗1_p=y⊗1_q (Defs. 2.1–2.3; Lemma 4.2) makes states with repeated coordinates the same state, e.g., (a) and (a,a), or (a,b) and (a,b,a,b). The paper's conclusion explicitly concedes that the physical meaningfulness of d_V is 'the only questionable point.' This is load-bearing: every controllability, observability, and DDP theorem in §5–§6 is stated on equivalence classes. If a physical system distinguishes two synchronized agents/generators from one, the theorems describe an artificial quotient. Please add an invariance/calibration test or a precise domain restriction.
  2. [Sec. III.D, Theorem 3.22] The stability theorem is not proved. The proof states that for sufficiently large dwell time ‖x̄_{t_k+}‖−‖x̄_{t_{k+1}-}‖>δ_{t_k}, but this inequality is not derived from Lyapunov condition (31), and no explicit bound on ∆_m is given in terms of L, the Lyapunov decay rate, or the jump bound. Uniform boundedness of δ_{t_k} is asserted, not shown. Either provide a quantitative dwell-time theorem or state the result as a heuristic. Note also the inconsistent use of ∆_m/∆_d.
  3. [Sec. V.B, Prop. 5.7] The proposition uses 'partly controllable on Ω_k \ Ω_{k+1}^⊥' without defining this notion, and without formalizing the chain condition or the concatenation of controls. The proof is two sentences and only asserts the conclusion. This undefined notion is also used in Proposition 5.11. The paper needs a precise definition and a proof of the sufficiency of the chain condition.
  4. [Sec. V.C, Prop. 5.10] Observability of a dimension-varying system is never defined. Proposition 5.10 then asserts that the system is 'completely observable iff it is completely observable on all modes.' This equivalence is nontrivial for switched systems and is not automatic if switching provides additional information. A formal definition of observability on Ω and a proof are required.
  5. [Sec. VI.C, Prop. 6.2] The DDP equivalence is asserted with a projection argument that is not sufficient. The statement 'ξ−Π_{n_k}^n ξ is orthogonal to R^{n_k}, so it will not affect the states of Σ_k' ignores that the lifted state evolves in R^n (n=lcm(n_k)); an orthogonal component can alter future modes or the output after lifting. A proof using invariant distributions—or a counterexample—is needed.
  6. [Abstract vs. full text] The abstract promises a 'generator-removal-and-reconnection scenario in a three-machine power system' and a finite-schedule dwell-time consistency check. No such case study appears in the body; Examples 3.23 and 6.1 are low-dimensional linear examples. Either include the advertised case study or revise the abstract.
minor comments (4)
  1. [Notation] F∞ should be R∞; Ω_0 in Prop. 5.11 is undefined; the orientation of Π_n^m is used inconsistently (e.g., Eq. (7) vs. Example 3.23); Eq. (55) refers to 'Σ described in (55)' but should probably be (54).
  2. [Typos] There are numerous typos: 'beaning', 'aircreft', 'nentwork', 'subspact', 'conclusing', 'Kharithnov', 'sting theory'.
  3. [References] References [17] and [22] appear to be the same paper ('Stability of switched linear differential systems') with different author transcriptions; please check.
  4. [Example 3.23] The numerical claim ∥Π_2^4∥_V=2 is not consistent with the operator norm formula (24) and the preceding computation; please verify the projection notation and the numerical value.

Circularity Check

0 steps flagged

No significant circularity: the Ω construction is explicitly defined in the paper and the control theorems follow from those definitions; self-citations supply supporting algebra, not fitted predictions.

full rationale

The central object Ω is not imported as an unproved premise: Definitions 2.1–2.3 explicitly define x±y, ⟨x,y⟩_V, ∥x∥_V, d_V, the equivalence x↔y iff d_V(x,y)=0, and the quotient Ω=R^∞/↔. The main topological claims (Proposition 3.2, Theorem 3.4) are proved directly from this definition, e.g. d_V(x̄,ȳ)=n^{-1/2}d(x,y) on R^n, so they are consequences rather than assumptions. The control-theoretic statements (controllability Prop. 5.7, observability Prop. 5.10, stabilization Prop. 5.11, DDP Prop. 6.2) are derived inside the explicitly stated Ω-framework; they do not fit parameters to data and then relabel the fit as a prediction. The paper does rely on the author's earlier works — 'This section is based on [6], [7], [10]' and 'We refer to [6], [10] for details of all above claims' — but those citations concern parameter-free algebraic constructions (semi-tensor product, projections, operator norm formula (24)) with published proofs, and the needed definitions are restated here. The conclusion's caveat, 'The only questionable point remains is: Is the distance physically meaningful?', is an external-validity limitation about whether repetition equivalence matches physical systems, not a circular derivation. No quoted equation reduces to its own input by construction, and no self-citation is used to forbid alternative frameworks. Thus the circularity score is low.

Axiom & Free-Parameter Ledger

3 free parameters · 5 axioms · 2 invented entities

The ledger shows the framework rests on two ad hoc choices (the repetition equivalence and the weighted semi-tensor scaling), on accepted prior results from the same author, and on unquantified assumptions about Lipschitz switching and dwell times. No empirical parameters are fitted; the central cost is the physical meaningfulness of d_V and the completeness of the deferred proofs.

free parameters (3)
  • DK-STP weighting factor n/t = n/t where t = lcm(n,p)
    Chosen by hand in Definition 2.6 to make A⋉B reduce to conventional AB for compatible dimensions; all matrix operations and operator norms inherit this scaling.
  • Projection weighting factor m/t = m/t in Π_n^m
    Chosen so that the projection is least-squares with respect to d_V; this factor defines the cross-dimensional metric geometry and the nearest-jump transitions used throughout.
  • Lipschitz switching constant L (or function φ) = unspecified
    Assumed in Definition 3.14 and equation (23) to bound jump growth; stability and dwell-time results depend on its existence, but no value or construction is given.
axioms (5)
  • domain assumption Prior cross-dimensional framework from [6], [7], [10] is correct, including inner product, projection formula (7), DK-STP properties, and operator norm formula (24).
    Section II states 'This section is based on [6],[7],[10]' and several claims are not re-proved; all later control theorems build on this foundation.
  • ad hoc to paper x↔y iff x⊗1_p = y⊗1_q captures 'same state across dimensions'; zero d_V is a meaningful state identity.
    Defined in Definition 2.3 and used throughout; the paper's conclusion explicitly says the physical meaningfulness of d_V remains questionable.
  • domain assumption Every switching satisfies a Lipschitz norm bound ||x(t+)||_V ≤ L||x(t-)||_V and equation (23) holds.
    Definition 3.14 assumes this for all switching moments; it is load-bearing for the stability and dwell-time arguments.
  • domain assumption There exists a common Lyapunov function L satisfying (31), and jump gaps δ_{t_k} are uniformly bounded.
    Theorem 3.22 assumes these to conclude stability for large enough dwell time; no construction or bound is provided.
  • domain assumption Minimum dwell time Δ_m ≥ ε > 0 and dwell time can be made large enough to dominate jump increases.
    Used in Definition 3.20 and Theorem 3.22; the proof only states 'as long as δ_{t_k} uniformly bounded' without a quantitative construction.
invented entities (2)
  • Ω = R^∞/↔ no independent evidence
    purpose: Universal state space for dimension-varying systems; each mode's R^{n_i} embeds as Ω_{n_i}.
    A quotient construction introduced by the paper. No external falsifiable prediction is made beyond the mathematical definitions; its utility is asserted through examples.
  • Cross-dimensional metric d_V and inner product ⟨·,·⟩_V no independent evidence
    purpose: Measure distances between states of different dimensions and define switching gaps and dwell-time bounds.
    The metric is defined by lifting to an lcm dimension and scaling. The paper itself calls its physical meaning into question in the concluding remarks.

reviewed 2026-08-03 · how reviews work

0 comments
Cite this review

Pith. "Pith review of On Dimension-Varying Control Systems: A Universal State Space Approach." pith.science (2026). https://pith.science/paper/2INPFSXG

@misc{pith2026260114839,
  author       = {Pith},
  title        = {Pith review of: On Dimension-Varying Control Systems: A Universal State Space Approach},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2INPFSXG}},
  note         = {Machine review of arXiv:2601.14839}
}
Share X Bluesky LinkedIn Reddit HN
abstract

This paper develops a unified framework for the analysis and design of dimension-varying control systems by constructing an intrinsic quotient state space, $\Omega$. A significant challenge in non-fixed-dimensional systems is the lack of a common metric space that enables comparison of states across dimensions without relying on arbitrary external embeddings. To address this, we propose a cross-dimensional pseudo-metric $d_{\mathcal{V}}$ on $\mathbb{R}^{\infty}$ and derive $\Omega$ by identifying zero-distance representatives. We demonstrate that $\Omega$ preserves the essential topological and metric geometry of Euclidean space, providing the necessary foundation to extend fundamental control notions to the dimension-varying case. Specifically, we establish conditions for controllability, observability and stabilizability, and we address the complexities of Lipschitz switching and disturbance decoupling within this common space. The framework is further extended to hierarchical dimension-varying networks. The practical utility of the results is illustrated through a generator-removal-and-reconnection scenario in a three-machine power system. This case study demonstrates the use of translated representatives and projection/lift benchmarks, quantifies event-wise $d_{\mathcal{V}}$-gaps, and provides a finite-schedule dwell-time consistency check to validate the system's structural transitions.

Figures

Figures reproduced from arXiv: 2601.14839 by Daizhan Cheng, Feng Liu, Tielong Shen, Xiaoming Hu.

Figure 2
Figure 2. Figure 2: Fig.2.1 Projection [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Fig.3.1 Comparing [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 3
Figure 3. Figure 3: Fig.3.2 A Triangle in [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Fig.4.1 Sub-Lattice as a State Space [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 4
Figure 4. Figure 4: Fig.4.2 A Continuous Trajectory in [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Fig.5.1 Norm of Trajectories [PITH_FULL_IMAGE:figures/full_fig_p013_5.png] view at source ↗
Figure 5
Figure 5. Figure 5: Fig.5.2 A Hierarchical Network [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

32 extracted references · 1 canonical work pages

  1. [1]

    Abbott,Set, Lattices and Boolean Algebras, Allun & Bacon, Boston, 1969

    J.C. Abbott,Set, Lattices and Boolean Algebras, Allun & Bacon, Boston, 1969

  2. [2]

    Boothby,An Introduction to Differential Manifold and Riemannian Geometry, Academic Press, Orlando, 1986

    W. Boothby,An Introduction to Differential Manifold and Riemannian Geometry, Academic Press, Orlando, 1986

  3. [3]

    Burris, H.P

    S. Burris, H.P. Sankappanavar,A Course in Universal Algebra, Springer, New York, 1981

  4. [4]

    Casarin, M

    R. Casarin, M. Iacopini, X. He, et al. Bayesian markov-switching tensor regression fo time-varying networks,J. American Stat. Assoc., 119(545), 109-121, 2024

  5. [5]

    J.C. Chan, G. Koop, R. Leon-Gonzalez, et al. Time varying dimension models,J Busin. Econ. Statis., 30(2010-523), 358-367

  6. [6]

    Cheng,From Dimension-Free Matrix Theory to Cross-Dimensional Dynamic Systems, Elsevier, London, 2019

    D. Cheng,From Dimension-Free Matrix Theory to Cross-Dimensional Dynamic Systems, Elsevier, London, 2019

  7. [7]

    Cheng, On equivalence of matrices,Asian J

    D. Cheng, On equivalence of matrices,Asian J. Mathematics, V ol. 23, No. 2, 257-348, 2019

  8. [8]

    Cheng, Z

    D. Cheng, Z. Xu, T. Shen, Equivalence-based model of dimension varying linear systems,IEEE Trans. Autom. Control, V ol. 65, No. 12, 5444-5449, 2020

  9. [9]

    Cheng, Z

    D. Cheng, Z. Ji,Lecture Notes in Semi-Tensor Product of Matrices, Vol. 4, Finite and Dimension-Free Dynamic Systems, Science Press, Beijing, 2023 (in Chinese)

  10. [10]

    Cheng, Cross-dimensional mathematics–A foun- dation for semi-tensor prodact/semi-tensor addition, htti//engine.scichina.com/doi/10.1007/s11425-024-2528-4

    D. Cheng, Cross-dimensional mathematics–A foun- dation for semi-tensor prodact/semi-tensor addition, htti//engine.scichina.com/doi/10.1007/s11425-024-2528-4

  11. [11]

    X. Feng, J. Liu, F.C. Lee, Impedance specifications for stable DC distributed power systems,IEEE Trans. Power Electr., 17(2) 157-162, 2002

  12. [12]

    Giordano, Passivity preserving energy-aware design for multi- dimensional switched systems: Application to open multi-robot systems, Automatica, 181(2025), 112496

    P.R. Giordano, Passivity preserving energy-aware design for multi- dimensional switched systems: Application to open multi-robot systems, Automatica, 181(2025), 112496

  13. [13]

    Hespanha, A.Sl Morse, Switching between stabilizing controllers, Automatica, 38(11), 1905-1917, 2002

    J.P. Hespanha, A.Sl Morse, Switching between stabilizing controllers, Automatica, 38(11), 1905-1917, 2002

  14. [14]

    Hungerford,Algebra, Springer-Verlag, New York, 1974

    T.W. Hungerford,Algebra, Springer-Verlag, New York, 1974

  15. [15]

    Isidori,Nonlinear Control Systems, 3rd Ed., Springer-Verlag, London, 1995

    A. Isidori,Nonlinear Control Systems, 3rd Ed., Springer-Verlag, London, 1995

  16. [16]

    P. Jiao, Y . Hao, J. Bin, Modeling and control of spacecraft formation based on impulsive switching with variable dimensions,Comput. Simul., V ol. 31, No. 6, 2014, 124-128

  17. [17]

    Jonathan, M

    C. Jonathan, M. Maldonado, P. Rapisarda, et al. Stability of switched linear differential systes,IEEE Trans. Aut. Contr., 59(8), 2038-2051, 2014

  18. [18]

    Kaku,Introduction to Supersting and M-Theory, 2nd Ed., Springer- Verlag, New York, 1999

    M. Kaku,Introduction to Supersting and M-Theory, 2nd Ed., Springer- Verlag, New York, 1999

  19. [19]

    Kelley,General Topology, Springer-Verlag, New York, 1955

    J.L. Kelley,General Topology, Springer-Verlag, New York, 1955

  20. [20]

    Kelley, I

    J.L. Kelley, I. Namioka,Linear Topological Spaces, Springer-Verlag, New York, 1963

  21. [21]

    Machowski, J.W

    J. Machowski, J.W. Bialek, J.R. Bumby,Power System Dynamics and Stability, John Wiley and Sons, Inc., Chichester, 1997

  22. [22]

    Myo-Maldonado, P

    J.C. Myo-Maldonado, P. Rapisarda, P. Rocha, Stability of switched linear differential systems,IEEE Trans. Aut. Contr., 59(8), 2038-2051, 2014

  23. [23]

    Minnichelli, J.J

    R.J. Minnichelli, J.J. Anagnost, C.A. Desoer, An elementary proof of Kharitonov’s theorem with extensions,IEEE Trans. Aut. Contr., 34(9), 995-998, 1989

  24. [24]

    Pakniyat, P.E

    A. Pakniyat, P.E. Caies, Hybrid optimal control of an electric vehicle with a dual-planetary transmissions,Nonlinear Systems, V ol. 25, 263- 282,2017

  25. [25]

    Restrepo, C

    E. Restrepo, C. Secchi, P.R. Giordano, Passivity preserving energy-aware design for multi-dimensional switched systems,Automatica, 181(2025) 112496

  26. [26]

    Song, D.F

    T. Song, D.F. Li, Hybrid time-varying frequency response function estimates using multiple sets of basis functions,IEEE Trans. Instrum. Meas., 66(2), 263-279, 2017

  27. [27]

    Taylor, D.C

    A.E. Taylor, D.C. Lay,Introduction to Functional Analysis, John Wiley & Sons, New York, 1980

  28. [28]

    Temporelli, M

    R. Temporelli, M. Boisvert, B. Micheau, Accurate clutch slip controllers during vehicle steady and acceleration states,IEEE/ASME Trans. Mecha- tronics, 23(5), 2078-2089, 2018

  29. [29]

    Verriest, Multi-mode multi-dimensional systems,Proc

    E.I. Verriest, Multi-mode multi-dimensional systems,Proc. 17th Int. Symp. Math. Theory Network Sys., 1268-1274, 2006

  30. [30]

    Verriest, Canonical forms for pseudo-continuous multi-mode multi- dimensional systems (M 3D) with conservation laws, inMathemamati- cal System Theory, 449-469, CreateSpace, 2013

    E.I. Verriest, Canonical forms for pseudo-continuous multi-mode multi- dimensional systems (M 3D) with conservation laws, inMathemamati- cal System Theory, 449-469, CreateSpace, 2013

  31. [31]

    M. Xue, Y . Tang, W. Ren, et al. Stability of mult-dimensional switched systems with application to opem multi-agent systems,Automatica, 146(2022), 110644

  32. [32]

    H. Yang, B. Jiang, V . Cocquempot,Stabilization of Switched Nonlinear Systems with Unstable Modes, Chapter 4. Switched Nonlinear Systems with Varying states, Springer, Switzerland, 2014

This paper was first reviewed by deepseek-v4-flash on August 3, 2026.