Pith. sign in

REVIEW 3 major objections 3 minor 37 references

Convergence Behavior Analysis of Directed Signed Networks Subject to Nonidentical Topologies

T0 review · 3 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper claims that directed signed networks with two nonidentical interaction graphs are fully classified by sign-consistency: if the two graphs agree on signs, polarization or neutralization is decided by structural balance of their…

desk verdict The main convergence classification is plausible and useful, but as stated the headline theorem is formally false because the paper's own definition of polarization includes neutralization as a special case. read the letter →

arxiv 1908.04467 v3 pith:ZP7I5GJO submitted 2019-08-13 math.DS

Deyuan Meng , Mingjun Du , Jianqiang Liang This is my paper · ORCID
classification math.DS MSC 93D0593A1405C5034D20
keywords signednetworksnonidenticaltopologiesstructuralbalancepolarizationneutralizationdirecteddigraphsmixed-orderdynamicsM-matrixmethod
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 studies directed networks in which some agents follow first-order dynamics and others second-order, and in which each interaction is cooperative or antagonistic, so the topology is a signed digraph. The new element is that the agents use two different signed digraphs at the same time, one for position coupling and one for velocity coupling, rather than a single topology. The paper claims a complete classification: when the two signed digraphs are sign-consistent, the agents polarize into two opposing groups exactly when the union of the two digraphs is structurally balanced, and they neutralize to zero exactly when the union is structurally unbalanced. When the two digraphs are sign-inconsistent, the paper claims the network always neutralizes, no matter the structure. The theorems matter because they turn an apparently intricate mixed-order, two-topology problem into a single graph-theoretic condition, and they give explicit damping thresholds that guarantee the predicted outcome.

What carries the argument

The machinery is the Laplacian algebra of signed digraphs plus a block-coordinate Lyapunov analysis. The nonsingular transformation (4) rewrites the mixed-order system as a block matrix whose diagonal blocks involve $kI+L_{B_c}$ and whose coupling involves $L_{B_c}+L_{B_d}$. The sign-consistency property (Definition 1) ensures that $L_{B_c}+L_{B_d}=L_{B_c+B_d}$, so the union of the two topologies has a genuine Laplacian; structural balance then gives a diagonal sign matrix $D$ with $DL_B D=L_{|B|}$. For a strongly connected union, Lemma 3 supplies positive-definite Lyapunov matrices $W$ (balanced case) or $H$ (unbalanced case), and the Schur complement argument turns the requirement of a negative-definite derivative into the explicit gain bound $k>\mu$. For sign-inconsistent topologies that Laplacian identity fails, so the paper factorizes $L_{B_c}+L_{B_d}$ from (11) as the product of an $M$-matrix and a nonnegative matrix, proves positive stability from Gershgorin circles and positivity of the determinant, and runs a second Lyapunov argument with threshold $k>\zeta$.

What would settle it

Simulate system (2) on a concrete instance of two sign-consistent signed digraphs whose union is strongly connected and structurally balanced, choose $k>\mu$ from (7), and let the positions run; if the final $|x_i|$ do not all approach one common value $\theta$ while velocities decay, Theorem 1 fails. For the sign-inconsistent case, simulate any pair with strongly connected union and $k>\zeta$; observing persistent velocities or positions that do not decay to zero would refute Theorem 2.

Watch

Extended reading notes

Core claim

The central discovery is that the long-run behavior of system (3) is governed by one new property, sign-consistency, together with the classical notion of structural balance. Theorem 1 considers sign-consistent topologies: with a strongly connected union and any damping gain $k>\mu$ (where $\mu$ is defined by (7)), polarization occurs if and only if the union graph $G(B_c+B_d)$ is structurally balanced, with the explicit limit $\lim_{t\to\infty} x(t)=\{\nu^T D[x(0)+k^{-1}y(0)]\}D1_n$ and $\lim_{t\to\infty} y(t)=0$; neutralization occurs if and only if the union is structurally unbalanced. Theorem 2 considers sign-inconsistent topologies: with a strongly connected union and any damping gain $k>\zeta$ (defined by (12)), neutralization is always achieved. Together the two theorems assert that sign-inconsistency is not a nuisance to be smoothed over but a decisive feature: it forces collapse to zero, whereas sign-consistency lets the network read its fate from the structural balance of the union digraph.

Load-bearing premise

The classification requires the union of the two signed digraphs to be strongly connected, meaning every agent influences every other through at least one of the two topologies; if the union splits into disconnected pieces, the stated if-and-only-if results are not proven and may fail piecewise.

Editorial extensions

If this is right

  • A single scalar test, structural balance of the union digraph, completely decides polarized versus neutralized behavior for sign-consistent nonidentical topologies, for any large enough damping gain.
  • Sign-inconsistent pairs with a strongly connected union cannot polarize: for sufficiently large $k$ every trajectory decays to zero regardless of the underlying signed structure.
  • When polarization occurs, the final configuration is explicit: agents split according to the balanced partition $D1_n$, with the common magnitude determined by $\nu^T D[x(0)+k^{-1}y(0)]$.
  • Because single-integrator and double-integrator signed networks are the two extremes of the mixed-order model (2), the two theorems bridge those previously separate settings through the nonidentical-topology lens.
  • The explicit thresholds $\mu$ and $\zeta$ convert the qualitative classification into a checkable design rule: choose $k$ above the threshold and the claimed convergence is guaranteed.

Reading between the lines

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

  • A natural untested extension is whether the dichotomy survives for arbitrary small $k>0$; the paper proves the threshold only for the specific Lyapunov constructions, so there may be sign-consistent pairs where intermediate gains cause oscillatory or even divergent behavior outside the theorem's remit.
  • The disconnected-union case is left open; when the union splits into several strongly connected components, one would expect the polarization-iff-balanced classification to hold component-wise after removing the extra zero eigenvalues, but the paper does not establish this.
  • The M-matrix factorization (11) is likely to generalize to more than two nonidentical topologies or to switching topologies, suggesting that sign-inconsistency's collapse mechanism is a robust phenomenon rather than an artifact of the two-graph setup.
  • Because the sign-consistency condition is stated entrywise, a practical check is cheap: inspect every pair of nonzero weights in $B_c$ and $B_d$; if any pair has opposite signs, Theorem 2 applies and predicts neutralization.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 3 minor

Summary. The paper studies convergence of directed signed networks with mixed first- and second-order dynamics, where the communication topology is described by two signed digraphs G(Bc) and G(Bd) that may be sign-consistent or sign-inconsistent and have a strongly connected union. The main results are Theorem 1, which claims that under sign-consistent topologies and a sufficiently large damping gain k, polarization occurs if and only if the union graph is structurally balanced and neutralization occurs if and only if it is structurally unbalanced, together with an explicit limit formula for the polarized state; and Theorem 2, which claims that under sign-inconsistent topologies and sufficiently large k, neutralization always occurs. The proofs use a nonsingular transformation, a Lyapunov function with an M-matrix approach, and matrix properties delegated to prior work by the same authors.

Significance. If the classification were correct as stated, the paper would provide a complete convergence characterization for a broad class of mixed-order directed signed networks with nonidentical topologies, going beyond identical-topology results. The introduction of sign-consistency as a structural property of pairs of signed digraphs is a useful concept, and the explicit converged solution in (22) as well as the M-matrix treatment of sign-inconsistent topologies are valuable contributions. The paper also gives falsifiable predictions: the sign pattern of the two topologies determines whether the network polarizes or neutralizes, with explicit lower bounds on the damping gain. These strengths make the underlying analysis worth considering, but the validity of the headline theorem is compromised by the definitional issue discussed below.

major comments (3)
  1. [Section II (definition of polarization) and Theorem 1] The iff classification in Theorem 1 is false under the stated definitions. Polarization is defined by lim_{t→∞} |x_i(t)| = θ and lim_{t→∞} y_i(t) = 0 with θ ≥ 0. Since neutralization requires lim_{t→∞} x_i(t) = 0 and lim_{t→∞} y_i(t) = 0, every neutralized trajectory also satisfies the polarization definition with θ = 0. Consequently, in any structurally unbalanced case where Theorem 1(2) and Case ii) of Appendix A establish neutralization, polarization is also achieved, contradicting the 'only if' direction of Theorem 1(1). Conversely, in a structurally balanced case, the converged solution (22) gives lim_{t→∞} x(t) = {ν^T D[x(0)+k^{-1}y(0)]} D 1_n. For initial conditions with ν^T D[x(0)+k^{-1}y(0)] = 0, the limit is zero, so neutralization is achieved despite structural balance, contradicting the 'only if' direction of Theorem 1(2). A concrete example is n = 2, Bc = Bd = [[0,1],[1,0]], D = I, ν = [1/2,1/2], x(0) = [1,-1]^T, y(0) = [0,0]^T. To repair the theorem, polarization should be defined with θ > 0 (excluding the zero limit), and the statements must be qualified to hold for generic initial conditions or for initial conditions with a nonzero conserved quantity, since the zero-conserved-quantity case inevitably produces neutralization even in balanced networks.
  2. [Appendix A, Proof of Theorem 1] The proof of the necessity directions is not adequate. The text states only that 'the necessity results of this theorem follow directly by the mutually exclusive relationship between the structural balance and unbalance of G(B)'. This is not a valid proof of the two 'if and only if' claims, especially because the definitions of polarization and neutralization overlap when θ = 0. After the definitional issue is fixed, the necessity arguments must be supplied explicitly: for balanced graphs one must rule out neutralization (except for the zero-measure initial-condition set), and for unbalanced graphs one must rule out nontrivial polarization.
  3. [Lemma 5 and its proof in Appendix B] The proof of Lemma 5 relies on the positive stability of Ξ = I - (L_{Bc+ + Bd+} + Δ|Bc- + Bd-|)^{-1}(Bc- + Bd-), which is asserted to follow from '[27, Lemma 4.2 and Corollary 4.2]' without reproducing the argument. Since this property is load-bearing for the conclusion that L_{Bc} + L_{Bd} is positive stable and hence for Theorem 2, the proof should either state the exact result from [27] or give a self-contained derivation. The same applies to the existence of W and H in Lemma 3, which is delegated to [27, Theorems 4.1 and 4.2].
minor comments (3)
  1. [Section III, Eq. (7) and (12)] The thresholds μ and ζ are defined for any δ > 1, so they are not single constants but families of constants indexed by δ. The theorems state 'let k > μ be selected for any δ > 1', which is logically ambiguous; it would be clearer to say 'for any δ > 1, let k > μ(δ)' or to fix δ first.
  2. [Theorem 1, limit formula] The vector ν in the converged solution (22) is described only as satisfying ν^T(DL_BD) = 0 and ν^T 1_n = 1. It would be helpful to state explicitly that ν is the normalized left null vector of the Laplacian L_{|B|}, whose existence and uniqueness follow from strong connectivity.
  3. [Throughout] The paper does not include numerical simulations. While not required for a theoretical result, a small simulation illustrating the sign-consistent balanced case, the sign-consistent unbalanced case, and the sign-inconsistent case would help the reader see the practical meaning of the thresholds μ and ζ.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the convergence proofs use standard Lyapunov and M-matrix arguments, with only minor self-citation reliance on the first author's earlier lemmas and a non-circular formal defect in the polarization definition.

full rationale

The paper's convergence results are not obtained by fitting parameters to the target conclusion: the thresholds μ and ζ are explicit sufficient conditions computed from Lyapunov matrices W and H, and the proofs establish negativity of Lyapunov derivatives rather than assuming the classification. The main external input is Lemma 3, which cites the same first author's [27] for existence and uniqueness of positive definite solutions to Lyapunov equations under strong connectivity and structural balance. That prior work is published, parameter-free, and its assumptions do not include the mixed-order nonidentical-topology conclusion; it supplies standard linear-algebra facts, so the self-citation is real evidence rather than a circular premise. Lemma 2 and Definitions 1-2 are definitional but do not presuppose Theorem 1, and the sign-inconsistent case is handled by a separate M-matrix argument. One formal issue deserves note but is not circularity: with θ ≥ 0 in the definition of polarization, neutralization is a special case of polarization, so the 'iff' classification in Theorem 1 is not a valid dichotomy (balanced graphs with zero conserved quantity can neutralize, and unbalanced graphs also satisfy the polarization definition with θ=0). This is a correctness/stating defect in the theorem, not a derivation that reduces to its own inputs; the Lyapunov/M-matrix convergence analysis itself is independent of that classification.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The central claim rests on two pillars borrowed from the same authors' prior work: the Lyapunov-matrix existence results in [27] and the sign-consistency notion from [30]. The thresholds μ and ζ are derived, not fitted. No new physical entities are introduced; the virtual node in the proof of Lemma 4 is an auxiliary mathematical construction.

assumptions (3)
  • domain assumption For a strongly connected and structurally balanced signed digraph G(B) with B = Bc + Bd sign-consistent, there exists a unique positive definite W satisfying ED(LBc+LBd)DF trace condition (5); and for unbalanced, there exists H satisfying (6).
    Lemma 3 is stated without proof, referencing [27, Theorems 4.1-4.2] by the same first author; the existence of W and H is assumed as background.
  • domain assumption In the sign-inconsistent case, Ξ = I - (LBc++Bd+ + ∆|Bc-+Bd-|)^{-1}(Bc-+Bd-) is positive stable.
    Lemma 5 proof delegates this to '[27, Lemma 4.2 and Corollary 4.2]', also by the same first author; no proof is reproduced in this paper.
  • standard math Standard Lyapunov stability theory, Schur complement lemma, Geršgorin disc theorem, and M-matrix properties from Horn and Johnson.
    Used throughout Appendices A and B; taken as unproved textbook results.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Convergence Behavior Analysis of Directed Signed Networks Subject to Nonidentical Topologies." pith.science (2026). https://pith.science/paper/ZP7I5GJO

@misc{pith2026190804467,
  author       = {Pith},
  title        = {Pith review of: Convergence Behavior Analysis of Directed Signed Networks Subject to Nonidentical Topologies},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZP7I5GJO}},
  note         = {Machine review of arXiv:1908.04467}
}
read the original abstract

This paper addresses the behavior analysis problems for directed signed networks that involve cooperative-antagonistic interactions among agents. Of particular interest is to explore the convergence behaviors of directed signed networks with agents of mixed first-order and second-order dynamics. Further, the agents are subject to nonidentical topologies represented by two different signed digraphs that have a strongly connected union. It is shown that when considering signed networks subject to sign-consistent nonidentical topologies, polarization (respectively, neutralization) can be achieved if and only if the union of two signed digraphs is structurally balanced (respectively, unbalanced). By comparison, signed networks can always be guaranteed to become neutralized in the presence of sign-inconsistent nonidentical topologies.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 34 canonical work pages

  1. [27]

    Convergence analysis of directed signed netw orks via an M - matrix approach,

    D. Meng, “Convergence analysis of directed signed netw orks via an M - matrix approach,” International Journal of Control , vol. 91, no. 4, pp. 827–847, Apr. 2018

    work page 2018

  2. [1]

    A tutorial on modeling a nd analysis of dynamic social networks. Part II,

    A. V . Proskurnikov and R. Tempo, “A tutorial on modeling a nd analysis of dynamic social networks. Part II,” Annual Reviews in Control , vol. 45, pp. 166–190, 2018

    work page 2018

  3. [2]

    Consensus seeking in multi-agent systems under dynamically changing interaction topologies,

    W. Ren and R. W. Beard, “Consensus seeking in multi-agent systems under dynamically changing interaction topologies,” IEEE Transactions on Automatic Control , vol. 50, no. 5, pp. 655–661, May 2005

    work page 2005

  4. [3]

    Tracking control for multi-ag ent consensus with an active leader and variable topology,

    Y . Hong, J. Hu, and L. Gao, “Tracking control for multi-ag ent consensus with an active leader and variable topology,” Automatica, vol. 42, no. 7, pp. 1177–1182, Jul. 2006. 6

    work page 2006

  5. [4]

    Some necessary and sufficient c on- ditions for second-order consensus in multi-agent dynamic al systems,

    W. Y u, G. Chen, and M. Cao, “Some necessary and sufficient c on- ditions for second-order consensus in multi-agent dynamic al systems,” Automatica, vol. 46, no. 6, pp. 1089–1095, Jun. 2010

    work page 2010

  6. [5]

    On consensus algorithms d esign for double integrator dynamics,

    A. Abdessameud and A. Tayebi, “On consensus algorithms d esign for double integrator dynamics,” Automatica, vol. 49, no. 1, pp. 253–260, Jan. 2013

    work page 2013

  7. [6]

    Coordination of multiple agents with double-integrator dynamics under generalized intera ction topolo- gies,

    J. Qin, W. X. Zheng, and H. Gao, “Coordination of multiple agents with double-integrator dynamics under generalized intera ction topolo- gies,” IEEE Transactions on Systems, Man, and Cybernetics–Part B: Cybernetics, vol. 42, no. 1, pp. 44–57, Feb. 2012

    work page 2012

  8. [7]

    Coordination of multiagents interacti ng under independent position and velocity topologies,

    J. Qin and C. Y u, “Coordination of multiagents interacti ng under independent position and velocity topologies,” IEEE Transactions on Neural Networks and Learning Systems , vol. 24, no. 10, pp. 1588–1597, Oct. 2013

    work page 2013

Show all 37 references
  1. [8]

    Distributed PID cont rol for consensus of homogeneous and heterogeneous networks,

    D. A. B. Lombana and M. di Bernardo, “Distributed PID cont rol for consensus of homogeneous and heterogeneous networks,” IEEE Transactions on Control of Network Systems , vol. 2, no. 2, pp. 154– 163, Jun. 2015

    2015

  2. [9]

    Containment c ontrol for second-order multiagent systems communicating over he terogeneous networks,

    J. Qin, W. X. Zheng, H. Gao, Q. Ma, and W. Fu, “Containment c ontrol for second-order multiagent systems communicating over he terogeneous networks,” IEEE Transactions on Neural Networks and Learning Sys- tems, vol. 28, no. 9, pp. 2143–2155, Sept. 2017

    2017

  3. [10]

    Consensus control for mult i-agent networks with mixed undirected interactions,

    W. Niu, D. Meng, and X. Ding, “Consensus control for mult i-agent networks with mixed undirected interactions,” in Proceedings of the Chinese Intelligent Systems Conference , Xiamen, China, pp. 101–108, Oct. 22-23, 2016

    2016

  4. [11]

    Network-to-netwo rk control over heterogeneous topologies: A dynamic graph approach,

    D. Meng, W. Niu, X. Ding, and L. Zhao, “Network-to-netwo rk control over heterogeneous topologies: A dynamic graph approach,” IEEE Transactions on Systems, Man, and Cybernetics: Systems , to appear, DOI: 10.1109/TSMC.2018.2791576

    2018

  5. [12]

    Consensus problems on networks with antag onistic inter- actions,

    C. Altafini, “Consensus problems on networks with antag onistic inter- actions,” IEEE Transactions on Automatic Control , vol. 58, no. 4, pp. 935–946, Apr. 2013

    2013

  6. [13]

    On the consensus and biparti te consensus in high-order multi-agent dynamical systems with antagoni stic interac- tions,

    M. E. V alcher and P . Misra, “On the consensus and biparti te consensus in high-order multi-agent dynamical systems with antagoni stic interac- tions,” Systems and Control Letters , vol. 66, no. 1, pp. 94–103, Apr. 2014

    2014

  7. [14]

    A lifting approach to models of opinio n dynamics with antagonisms,

    J. M. Hendrickx, “A lifting approach to models of opinio n dynamics with antagonisms,” in Proceedings of the IEEE Conference on Decision and Control, Los Angeles, CA, USA, pp. 2118–2123, Dec. 15-17, 2014

    2014

  8. [15]

    Opinion dyn amics in social networks with hostile camps: Consensus vs. polariza tion,

    A. V . Proskurnikov, A. Matveev, and M. Cao, “Opinion dyn amics in social networks with hostile camps: Consensus vs. polariza tion,” IEEE Transactions on Automatic Control , vol. 61, no. 6, pp. 1524–1536, Jun. 2016

    2016

  9. [16]

    Bipartite consensus of multi-age nt systems over signed graphs: State feedback and output feedback control a pproaches,

    H. Zhang and J. Chen, “Bipartite consensus of multi-age nt systems over signed graphs: State feedback and output feedback control a pproaches,” International Journal of Robust and Nonlinear Control , vol. 27, pp. 3– 14, Jan. 2017

    2017

  10. [17]

    Adaptive finite-time biparti te consensus for second-order multi-agent systems with antagonistic in teractions,

    L. Zhao, Y . Jia, and J. Y u, “Adaptive finite-time biparti te consensus for second-order multi-agent systems with antagonistic in teractions,” Systems and Control Letters , vol. 102, pp. 22-31, Apr. 2017

    2017

  11. [18]

    Dynamic distributed control for networks wit h cooperative- antagonistic interactions,

    D. Meng, “Dynamic distributed control for networks wit h cooperative- antagonistic interactions,” IEEE Transactions on Automatic Control, vol. 63, no. 8, pp. 2311–2326, Aug. 2018

    2018

  12. [19]

    Bipartite tracking con sensus of linear multi-agent systems with a dynamic leader,

    G. Wen, H. Wang, X. Y u, and W. Y u, “Bipartite tracking con sensus of linear multi-agent systems with a dynamic leader,” IEEE Transactions on Circuits and System II: Express Briefs , vol. 65, no. 9, pp. 1204–1208, Sept. 2018

    2018

  13. [20]

    Sign-consensus of line ar multi-agent systems over signed directed graphs,

    Y . Jiang, H. Zhang, and J. Chen, “Sign-consensus of line ar multi-agent systems over signed directed graphs,” IEEE Transactions on Industrical Electronics, vol. 64, no. 6, pp. 5075–5083, Jun. 2017

    2017

  14. [21]

    Beh aviors of networks with antagonistic interactions and switching t opologies,

    Z. Meng, G. Shi, K. H. Johansson, M. Cao, and Y . Hong, “Beh aviors of networks with antagonistic interactions and switching t opologies,” Automatica, vol. 73, pp. 110–116, Nov. 2016

    2016

  15. [22]

    Exponenti al convergence of the discrete and continuous-time Altafini models,

    J. Liu, X. Chen, T. Basar, and M. A. Belabbas, “Exponenti al convergence of the discrete and continuous-time Altafini models,” IEEE Transactions on Automatic Control , vol. 62, no. 12, pp. 6168–6182, Dec. 2017

    2017

  16. [23]

    Opinion behavior analysi s in social networks under the influence of coopetitive media,

    D. Xue, S. Hirche, and M. Cao, “Opinion behavior analysi s in social networks under the influence of coopetitive media,” IEEE Transactions on Network Science and Engineering , to appear, DOI: 10.1109/TNSE.2019.2894565

    2019

  17. [24]

    Emergent collective behaviors on coopetition networks,

    J. Hu and W. X. Zheng, “Emergent collective behaviors on coopetition networks,” Physics Letters A , vol. 378, nos. 26-27, pp. 1787–1796, May 2014

    2014

  18. [25]

    Interval bipartite consensu s of networked agents associated with signed digraphs,

    D. Meng, M. Du, and Y . Jia, “Interval bipartite consensu s of networked agents associated with signed digraphs,” IEEE Transactions on Auto- matic Control, vol. 61, no. 12, pp. 3755–3770, Dec. 2016

    2016

  19. [26]

    Structural balance and opinion separation in trust-mistrust social networks,

    W. Xia, M. Cao, and K. H. Johansson, “Structural balance and opinion separation in trust-mistrust social networks,” IEEE Transactions on Control of Network Systems , vol. 3, no. 1, pp. 46–56, Mar. 2016

    2016

  20. [28]

    Bipartite flocking fo r multi- agent systems,

    M. C. Fan, H. T. Zhang, and M. Wang, “Bipartite flocking fo r multi- agent systems,” Communications in Nonlinear Science and Numerical Simulation, vol. 19, no. 9, pp. 3313–3322, Sept. 2014

    2014

  21. [29]

    Bipartite containment tracking of signed net works,

    D. Meng, “Bipartite containment tracking of signed net works,” Auto- matica, vol. 79, pp. 282–289, May 2017

    2017

  22. [30]

    Bipartite consensus for second-o rder multi- agent systems over nonidentical signed graphs,

    J. Liang and D. Meng, “Bipartite consensus for second-o rder multi- agent systems over nonidentical signed graphs,” in Proceedings of the 7th IF AC W orkshop on Distributed Estimation and Control in Networked Systems, Groningen, Netherlands, pp. 301–306, Aug. 27-28, 2018

    2018

  23. [31]

    R. A. Horn and C. R. Johnson, Topics in Matrix Analysis . Cambridge: Cambridge University Press, 1991

    1991

  24. [32]

    Consensus of heterogeneo us multi- agent systems,

    Y . Zheng, Y . Zhu, and L. Wang, “Consensus of heterogeneo us multi- agent systems,” IET Control Theory and Applications , vol. 5, no. 16, pp. 1881–1888, Nov. 2011

    2011

  25. [33]

    Consensus of he terogeneous first- and second-order multi-agent systems with directed c ommunication topologies,

    Y . Feng, S. Xu, F. L. Lewis, and B. Zhang, “Consensus of he terogeneous first- and second-order multi-agent systems with directed c ommunication topologies,” International Journal of Robust and Nonlinear Control , vol. 25, no. 3, pp. 362–375, Feb. 2015

    2015

  26. [34]

    Group con sensus control for heterogeneous multi-agent systems with fixed an d switching topologies,

    G. Wen, J. Huang, C. Wang, Z. Chen, and Z. Peng, “Group con sensus control for heterogeneous multi-agent systems with fixed an d switching topologies,” International Journal of Control , vol. 89, no. 2, pp. 259– 269, Feb. 2016

    2016

  27. [35]

    R. A. Horn and C. R. Johnson, Matrix Analysis. Cambridge: Cambridge University Press, 1985. APPENDIX A: P ROOF OF THEOREM 1 Proof of Theorem 1: Since G (Bc) and G ( Bd) are sign- consistent, we use Lemma 2 to denote the union G (B) of them with B = Bc + Bd. Thus, the necessity...

    1985

  28. [36]

    This ensures Bc− + Bd− ⁄= 0, based on which we can construct a nonnegative matrix as A = [ 0 0 ∆ |Bc− +Bd− |1n Bc+ + Bd+ ] ≥ 0

    It thus yields bc− ij + bd− ij < 0. This ensures Bc− + Bd− ⁄= 0, based on which we can construct a nonnegative matrix as A = [ 0 0 ∆ |Bc− +Bd− |1n Bc+ + Bd+ ] ≥ 0. (26) Denote A ≜ [aij] ∈ R(n+1)× (n+1). We can define an unsigned digraph G ( A ) = ( V, E, A ) , in which we set V...

  29. [37]

    With this property, we first prove that the unsigned digraph G ( A ) contains a spanning tree, and then that LBc++Bd+ + ∆ |Bc− +Bd− | is an M -matrix to complete this proof. We know from (26) that G ( A ) is composed of the unsigned digraph G ( Bc+ + Bd+) and the vertex v0 with...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.