Pith. sign in

REVIEW 6 major objections 5 minor 29 references

Effective Fixed-Time Control for Constrained Nonlinear System

T0 review · 6 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 An adaptive fixed-time backstepping design with four event-triggered update rules aims to keep non-strict full-state-constrained nonlinear systems stable in preset time, with fewer controller updates and tracking error bounded by design…

desk verdict A well-motivated incremental framework, but load-bearing errors in the state transformation, the event-triggered proof, and the simulation parameters undermine the central theoretical claims as written. read the letter →

arxiv 2505.20870 v1 pith:T7J2VPK5 submitted 2025-05-27 eess.SY cs.SY

Chenglin Gong , Ziming Wang , Guanxuan Jiang , Xin Wang , Yiding Ji This is my paper
classification eess.SYcs.SY
keywords adaptivefixed-timecontrolevent-triggeredself-triggeredfull-stateconstraintsnonlinearmappingtransformationRBFneuralnetworksbacksteppingsemi-globalpracticalstability
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 claims that a nonlinear system whose states must stay inside asymmetric intervals can be tracked by an adaptive fixed-time backstepping controller that never lets the states leave their bounds. The key move is a one-to-one logarithmic mapping that converts the constrained coordinates into free coordinates, so the usual recursive backstepping construction applies. The authors add four event-triggered update rules—fixed-threshold, relative-threshold, switched-threshold, and self-triggered—and prove that under each rule the closed loop is semi-globally practically fixed-time stable, meaning all signals reach a small residual set within a fixed time for any bounded initial set, with a settling-time bound independent of initial conditions. Their simulations report that all four strategies cut the number of control updates substantially while preserving tracking accuracy. If the proof holds, the framework offers a single comparison point for choosing an event-triggering strategy under state constraints.

What carries the argument

The central object is the asymmetric logarithmic state transformation (3), $w_i = \log((\rho_{s1}+x_i)/(\rho_{s1}-x_i))$, which the paper treats as a one-to-one map from the constraint interval $(-\rho_{s1},\rho_{s2})$ to the real line and differentiates to obtain $\dot w_i = \Delta_i \dot x_i$ with $\Delta_i = (e^{w_i}+e^{-w_i}+2)/(\rho_{s1}+\rho_{s2})$. This map is meant to convert the constrained system (1) into the pure-feedback form (5), on which the backstepping Lyapunov steps are built. The supporting machinery is a bank of RBF neural networks that approximate the unknown lumped functions $U_i$, the four event-triggering conditions that freeze the actual control $g(t)=d(t_j)$ between updates, and Lemma 2's semi-global practical fixed-time stability criterion, which converts the differential inequality $\dot V \le -aV^q - bV^p + c$ into a settling time $T_{\max} = 1/(aI(q-1)) + 1/(bI(1-p))$ independent of initial conditions.

What would settle it

Differentiate the printed $w_i = \log((\rho_{s1}+x_i)/(\rho_{s1}-x_i))$ with respect to $x_i$: the result is $2\rho_{s1}/(\rho_{s1}^2-x_i^2)$, not the printed $\Delta_i$ unless $\rho_{s1}=\rho_{s2}$. A concrete check is to rerun the paper's simulation with $\rho_{s1}=1,\rho_{s2}=2$ using the correct derivative; if the claimed tracking bound and fixed-time settling are lost, Theorem 1 as written is falsified.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem 1: the closed loop composed of system (1), the virtual control laws, the actual controller (11), and the adaptive law (12) is semi-globally practically fixed-time stable under any of the four event-triggered strategies. Specifically, there are gains $a,b>0$, $0<I<1$, $q>1$, $p<1$, and $c>0$ such that the Lyapunov function obeys $\dot V \le -aV^q - bV^p + c$, so every signal enters a set of radius $\min\{(c/((1-I)a))^{1/q},\, (c/((1-I)b))^{1/p}\}$ within time $T_{\max}=1/(aI(q-1)) + 1/(bI(1-p))$, independent of initial conditions, and the tracking error satisfies $|y-y_m|\le 2(c/((1-I)a))^{1/(2q)}$. The four trigger strategies differ only in the constant $\Lambda$ in the Lyapunov derivative—fixed: $0.2785R\Phi$; relative and self-triggered: $0.557R\Phi$; switched: $0.8355R\Phi$—so a single stability proof covers all of them.

Load-bearing premise

The load-bearing premise is that the logarithmic coordinate change in Eq. (3) is a one-to-one map from $(-\rho_{s1},\rho_{s2})$ to the real line with derivative $\Delta_i=(e^{w_i}+e^{-w_i}+2)/(\rho_{s1}+\rho_{s2})$; taken literally, that derivative is consistent only when $\rho_{s1}=\rho_{s2}$, and the transformed system (5) used in every later step does not follow for the asymmetric constraints claimed.

Editorial extensions

If this is right

  • Under any of the four strategies, Theorem 1 asserts the closed loop is semi-globally practically fixed-time stable, with all signals bounded and entering a small residual set by time $T_{\max}$, independent of initial conditions.
  • The tracking error obeys $|y-y_m| \le 2(c/((1-I)a))^{1/(2q)}$, so a designer can shrink the guaranteed error range by choosing the design gains and thresholds.
  • The four trigger rules differ only in the Lyapunov constant $\Lambda$, so a single stability proof covers fixed, relative, switched, and self-triggered communication policies.
  • The self-triggered strategy needs no continuous monitoring of state variables; it computes the next trigger time from the current control magnitude and its rate of change.
  • Simulation over 20,000 test cycles reports trigger counts of 439, 565, 151+347, and 798 for the fixed, relative, switched, and self-triggered strategies, with relative-threshold tracking the most accurate.

Reading between the lines

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

  • Beyond the paper: if the derivative in Eq. (3) is corrected to $2\rho_{s1}/(\rho_{s1}^2-x_i^2)$ for asymmetric intervals, or the map is redefined to respect $\rho_{s2}$ explicitly, the same backstepping structure would likely survive, but the constants in Theorem 1 and Assumption 1 would need re-derivation.
  • Beyond the paper: the reported trigger counts can be read as a two-dimensional trade-off curve between tracking error and communication cost, letting a designer choose among the four strategies from an accuracy budget.
  • Beyond the paper: the self-triggered update rule (23) uses a bound on $\dot d(t)$; implementing it with unmeasured disturbances would require an extra observer or Lipschitz assumption to compute that bound in real time.
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

6 major / 5 minor

Summary. The paper proposes an adaptive fixed-time backstepping controller for non-strict full-state-constrained nonlinear systems, using a logarithmic state transformation and RBFNN approximation, and combines it with four event-triggered strategies (fixed, relative, switched, self-triggered). The main claim (Theorem 1) is that the closed-loop system is semi-globally practically fixed-time stable with a settling time independent of initial conditions, tracking error bounded by an explicit constant, and reduced trigger frequency. A two-state numerical example is presented as validation. The general architecture is standard for this literature, but several load-bearing steps in the transformation, the event-triggered Lyapunov analysis, and the stability proof are internally inconsistent as written.

Significance. If the claims were rigorously established, the paper would provide a useful comparison of four event-triggered strategies in a state-constrained fixed-time setting, and the proposed multi-threshold framework could be of practical interest for communication-limited control. The authors also deserve credit for clearly separating the four triggering mechanisms and for giving explicit settling-time and tracking-error formulas. However, the current manuscript does not make these results reliable: the state-transformation equation is inconsistent with its printed definition, the relative-threshold proof rests on a false sign assertion, the algebraic derivation of the central Lyapunov inequality is not reproducible, and the stability proof invokes an unproved bound on the adaptive parameter error. The simulation also violates the p<1 hypothesis of the fixed-time lemma. No machine-checked proofs, reproducible code, or data artifacts are provided, so the numerical claims cannot be independently verified.

major comments (6)
  1. [§III-A, Eq. (3), Assumption 1] As printed, w_i = log((rho_{s1}+x_i)/(rho_{s1}-x_i)) has derivative 2 rho_{s1}/(rho_{s1}^2-x_i^2), not Delta_i = (e^{w_i}+e^{-w_i}+2)/(rho_{s1}+rho_{s2}). The printed Delta_i corresponds instead to the asymmetric transformation w_i = log((rho_{s1}+x_i)/(rho_{s2}-x_i)). Since the transformed system (5) is built on this Delta_i, all subsequent backstepping and Lyapunov steps do not follow from the equations as stated. In addition, Assumption 1 postulates a fixed bound |Delta_i(w_i)| <= R^* although Delta_i grows unbounded as |w_i| -> infinity; a compact domain or an alternative treatment is required before R^* can be used as a constant in the Lyapunov bounds.
  2. [§III-B, Eqs. (16)-(18)] The assertion immediately after Eq. (16) that 'from (16) we get d(t) < 0' is false. Both alpha_n tanh(z_n alpha_n/Phi) and bar_vartheta_1 tanh(z_n bar_vartheta_1/Phi) have the sign of z_n, so d(t) = -(1+theta)(...) has the opposite sign of z_n and is positive whenever z_n < 0. The inequality d(t)/(1+beta_1(t)theta) <= d(t)/(1+theta), used to obtain Eq. (18), is valid only when d(t) <= 0 and reverses for d(t) > 0. Therefore Eq. (18) is not derived, and the same gap propagates to the switched-threshold bound (21) and the self-triggered bound (24), which are claimed to follow 'similar to (18)'.
  3. [§III-C, Eqs. (27)-(28)] The displayed inequality (27) contains the term -pi(sum_o tilde_phi_o^2/(2epsilon_o))^q twice, and the terms +sum_o tau_o(tilde_phi_o^2/(2epsilon_o))^q and -sum_o tau_o(tilde_phi_o^2/(2epsilon_o))^q cancel immediately. Yet Eq. (28) reintroduces +sum_o tau_o(tilde_phi_o^2/(2epsilon_o))^q - sum_o tau_o tilde_phi_o^2/(2epsilon_o). As printed, the algebra from (27) to (28) is internally inconsistent, so the central Lyapunov inequality dot V <= -aV^q - bV^p + c1 is not established by the proof.
  4. [§III-C, after Eq. (28), before Eq. (29)] The proof introduces the assumption that there exists an unknown gamma_m with |tilde_phi_m| < gamma_m and uses it to classify cases and to define the constant c in (31)-(32). No mechanism in the adaptive law (12) or in the virtual control laws guarantees this bound a priori; boundedness of tilde_phi is precisely part of what Theorem 1 must prove. The case analysis is therefore circular unless a projection operator or an independent boundedness argument is supplied. The constant c1 also contains tilde_phi-dependent terms, so the reduction to Lemma 2 depends on this unproved premise.
  5. [§II-A, Eq. (1), and §III-A, Theorem 1] The plant is defined with output y = x_n, but the control objective is formulated through z_1 = w_1 - w_s, which tracks the first state x_1, and the simulation in Eq. (33) sets y = x_1. The final bound |y - y_m| in Theorem 1 is not connected to the output of the stated plant. As a result, the tracking-error claim is not well posed with respect to the model in Eq. (1).
  6. [§IV, Lemmas 2 and simulation parameters] Lemma 2 requires p < 1 for the fixed-time Lyapunov argument and the settling-time formula T_max = 1/(aI(q-1)) + 1/(bI(1-p)), but the simulation sets p = 1.5 with q = 1.05. With p > 1, the exponent 1-p is negative and the stated settling-time bound is not applicable. The numerical example therefore does not demonstrate the claimed fixed-time property under the hypotheses of the paper.
minor comments (5)
  1. [Appendix, Eq. (37)] In the Step i proof, the coupling term appears as z_i z_{i-1}, whereas the recursive structure and Eq. (8) require z_i z_{i+1}; please correct this typo.
  2. [§II-B, Lemma 1, Eq. (2)] The inequality is typeset with an unclear exponent involving r_3 that makes the statement difficult to parse; please reprint the standard Young inequality cleanly.
  3. [§III-A, Appendix] The ideal weights phi_o = ||H_o||^2 are used in the estimates but are not defined before Eq. (34); please define them in Section II-B.
  4. [§IV, around Table I] The claim that 20,000 test cycles represent a significant reduction in triggering frequency cannot be assessed without specifying the simulation horizon, sampling step, or the number of time steps; please include these details.
  5. [Throughout] There are several typographical errors, including 'Finantial support' in the footnote, 'Internaional Journal' in reference [17], and 'Y , Gao' in reference [19]; a careful proofread is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation is a standard backstepping/RBFNN Lyapunov argument, and the self-cited lemmas are independently checkable.

full rationale

The paper's derivation chain is not circular. The controller (11), adaptive law (12), and virtual control laws are constructed so that the Lyapunov function derivative satisfies an inequality of the form Vdot <= -aV^q - bV^p + c, and the resulting settling-time and tracking-error bounds are expressed in terms of the user-selected design constants a, b, I, and c rather than being fitted to the simulation outcome. The lemmas imported from the authors' prior work (Lemma 1 and Lemma 2 from [6]) are standard, parameter-free inequalities or stability criteria that are independently verifiable and do not assume Theorem 1; Lemma 3 from [22] is likewise an external, standard tanh-approximation inequality. The threshold constants 0.2785R*Phi, 0.557R*Phi, and 0.8355R*Phi come from applying the tanh approximation lemma to the event-triggered control terms, not from reverse-engineering the simulation results. The proof concerns identified in the review are correctness issues rather than circularity: the printed derivative in Eq. (3) does not match the printed definition of w_i, and the assertion 'from (16) we get d(t) < 0' is sign-sensitive and appears false when z_n < 0, which would undermine the relative/switched/self-triggered Lyapunov bounds. These are mathematical gaps, not cases where a prediction reduces to its own inputs by construction. Therefore the circularity score is 0.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

No new physical or ontological entities are introduced. The auxiliary systems L_i and the logarithmic mapping are mathematical transformations standard in constrained backstepping. The load-bearing inputs are the RBF approximation assumption, the uniform uncertainty bound R*, the unproved parameter-error bound gamma_m, and the validity of the transformed system, all of which are either external lemmas or ad hoc premises.

free parameters (4)
  • Backstepping gains k_{1,o}, k_{2,o}, tau_o, u_o, f_o, epsilon_o = Simulation uses k_{1,1}=800, k_{1,2}=18 or 19, tau_1=tau_2=10, f_1=6, f_2=3, u_1=1, U_2=1
    Chosen by hand with no tuning rule; the Lyapunov proof only requires positivity, while the tracking-error bound depends on their values.
  • Event-triggered threshold parameters theta, vartheta, vartheta_1, bar_vartheta, bar_vartheta_1, G, pi, Phi = Fixed: vartheta=5, Phi=900; relative: theta=0.1, bar_vartheta_1=15, vartheta_1=0.1; switched: G=80, vartheta=4…
    These are selected per strategy and enter the Lyapunov bound through Lambda; no method is given for choosing them, and the trigger counts depend directly on them.
  • Fixed-time exponents p and q = Simulation lists q=1.05 and p=1.5
    Lemma 2 requires p<1 and q>1; the printed p=1.5 violates the fixed-time stability condition, so the convergence-time formula does not apply to the simulated configuration.
  • Constraint bounds rho_{s1}, rho_{s2} = rho_{11}=1, rho_{12}=2, rho_{21}=8, rho_{22}=9
    User-specified state bounds that appear in the logarithmic mapping and in the derivative Delta_i; they are chosen by hand for the example.
assumptions (5)
  • domain assumption RBF neural networks uniformly approximate the composed unknown functions U_i(Z_i) on a relevant compact set.
    Used in the appendix to replace U_i by H_i^T Omega_i + vartheta_i with bounded error lambda_i; no explicit construction of the compact set or approximation bounds is provided.
  • domain assumption Assumption 1: the uncertainty term Delta_i(w_i) satisfies |Delta_i(w_i)| <= R* for an unknown constant R*.
    This uniform bound is required to absorb the event-triggering errors into the Lyapunov constant Lambda; the existence of such a bound is assumed, not established.
  • ad hoc to paper For each adaptive loop there exists an unknown gamma_m with |tilde_phi_m| < gamma_m.
    Introduced in the Theorem 1 proof after Eq. (28) to split cases and absorb parameter-error terms. Boundedness of tilde_phi is part of what the stability theorem must prove, so this premise is not independently established.
  • standard math Lemma 2 fixed-time stability criterion is valid and applicable to V_n.
    Imported from reference [6]; the paper does not prove it, and the simulation parameters later violate its p<1 hypothesis.
  • domain assumption The transformed system (5) is equivalent to the constrained system (1) under the one-to-one mapping (3).
    This equivalence is the backbone of the backstepping design. As printed, Eq. (3) is inconsistent with the stated constraints and with the derivative Delta_i, so the equivalence is not established.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Effective Fixed-Time Control for Constrained Nonlinear System." pith.science (2026). https://pith.science/paper/T7J2VPK5

@misc{pith2026250520870,
  author       = {Pith},
  title        = {Pith review of: Effective Fixed-Time Control for Constrained Nonlinear System},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T7J2VPK5}},
  note         = {Machine review of arXiv:2505.20870}
}
read the original abstract

In this paper, we tackle the state transformation problem in non-strict full state-constrained systems by introducing an adaptive fixed-time control method, utilizing a one-to-one asymmetric nonlinear mapping auxiliary system. Additionally, we develop a class of multi-threshold event-triggered control strategies that facilitate autonomous controller updates, substantially reducing communication resource consumption. Notably, the self-triggered strategy distinguishes itself from other strategies by obviating the need for continuous real-time monitoring of the controller's state variables. By accurately forecasting the subsequent activation instance, this strategy significantly optimizes the efficiency of the control system. Moreover, our theoretical analysis demonstrates that the semi-global practical fixed-time stability (SPFTS) criterion guarantees both tracking accuracy and closed-loop stability under state constraints, with convergence time independent of initial conditions. Finally, simulation results reveal that the proposed method significantly decreases the frequency of control command updates while maintaining tracking accuracy.

Figures

Figures reproduced from arXiv: 2505.20870 by the authors.

Figure 1
Figure 1. The consistency tracking performance involving signal [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. The change of the actual controller g under different ETC strategies, where (a), (b), (c), and (d) represent the fixed threshold, relative threshold, switched threshold and self-triggered strategies respectively. (a) (b) (c) (d) [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. The trigger interval for event triggering under different ETC strategies, where (a), (b), (c), and (d) represent the fixed threshold, relative threshold, [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

29 extracted references · 29 canonical work pages

  1. [6]

    Z. Wang, X. Wang and N. Pang, ”Adaptive Fixed-Time Control for Full State-Constrained Nonlinear Systems: Switched-Self-Triggered Case” IEEE Trans. Cir. Sys. II Exp. Briefs , vol. 71, no. 2, pp. 752-756, 2024

    work page 2024

  2. [19]

    Wang, Y , Gao, A

    Z. Wang, Y , Gao, A. I. Rikos, N. Pang, and Y . Ji, ”Fixed-Relative- Switched Threshold Strategies for Consensus Tracking Control of Non- linear Multiagent Systems,” the 19th IEEE International Conference on Control & Automation , 2025

    work page 2025

  3. [1]

    Krstic, I

    M. Krstic, I. Kanellakopoulos, and P. V . Kokotovic, Nonlinear and Adaptive Control Design . New York, NY , USA: Wiley, 1995

    work page 1995

  4. [2]

    M. Chen, S. S. Ge and B. Ren, ”Adaptive tracking control of uncertain MIMO nonlinear systems with input constraints”, Automatica, vol. 47, no. 3, pp. 452-465, 2011

    work page 2011

  5. [3]

    X. Wang, Z. Yin, Y . Lei, T. Huang, and J. Kurths, ”Secure consensus for switched multiagent systems under DoS attacks: Hybrid event-triggered and impulsive control approach,” IEEE Transactions on Cybernetics , 2025

    work page 2025

  6. [4]

    Wang, ”Hybrid Event-triggered Control of Nonlinear System with Full State Constraints and Disturbance,” 36th Chinese Control Decision Conference, pp

    Z. Wang, ”Hybrid Event-triggered Control of Nonlinear System with Full State Constraints and Disturbance,” 36th Chinese Control Decision Conference, pp. 2122-2127, 2024

    work page 2024

  7. [5]

    N. Pang, X. Wang, and Z. Wang, ”Event-triggered adaptive control of nonlinear systems with dynamic uncertainties: The switching threshold case,” IEEE Trans. Circuits Systems II Express Briefs , vol. 69, no. 8, pp. 3540-3544, 2022

    work page 2022

  8. [7]

    X. Wang, S. Zhang, H. Li, W. Zhang, H. Li and T. Huang, ”Neuroad- aptive Containment Control for Nonlinear Multiagent Systems With Input Saturation: An Event-Triggered Communication Approach,”IEEE Trans. on Systems, Man and Cybernetics: Systems , pp. 1-11, 2025

    work page 2025

Show all 29 references
  1. [8]

    Michalska and D

    H. Michalska and D. Q. Mayne, ”Robust receding horizon control of constrained nonlinear systems”, IEEE Transactions on Automatic Control, vol. 38, no. 11, pp. 1623-1633, 1993

    1993

  2. [9]

    Bemporad, ”Reference governor for constrained nonlinear systems”, IEEE Trans

    A. Bemporad, ”Reference governor for constrained nonlinear systems”, IEEE Trans. Autom. Control , vol. 43, no. 3, pp. 415-419, 1998

    1998

  3. [10]

    Y . J. Liu and S. C. Tong, ”Barrier Lyapunov Functions-based adaptive control for a class of nonlinear pure-feedback systems with full state constraints”, Automatica, vol. 64, pp. 70-75, 2016

    2016

  4. [11]

    Polyakov, D

    A. Polyakov, D. Efimov and W. Perruquetti, ”Finite-time and fixed- time stabilization: Implicit Lyapunov function approach”, Automatica, vol. 51, pp. 332-340, 2015

    2015

  5. [12]

    Z. Y . Zuo, Q. L. Han, B. Ning, X. H. Ge and X. M. Zhang, ”An Overview of Recent Advances in Fixed-Time Cooperative Control of Multiagent Systems”, IEEE Trans. Ind. Inform., vol. 14, no. 6, pp. 2322- 2334, 2018

    2018

  6. [13]

    Aldana-Lopez, D

    R. Aldana-Lopez, D. Gomez-Gutierrez, E. Jimenez-Rodriguez, J. D. Sanchez-Torres and M. Defoort, ”Enhancing the settling time estimation of a class of fixed-time stable systems”, Int. J. Robust Nonlinear Control, vol. 29, no. 12, pp. 4135-4148, Aug. 2019

    2019

  7. [14]

    Y . Liu, H. Li, Z. Zuo, X. Li and R. Lu, ”An Overview of Finite/Fixed- Time Control and its Application in Engineering Systems”, IEEE-CAA J. Autom. Sinica , 2022

    2022

  8. [15]

    W. P. M. H. Heemels, K. H. Johansson and P. Tabuada, ”An intro- duction to event-triggered and self-triggered control”, 2012 IEEE 51st IEEE Conference on Decision and Control (CDC) , 2012

    2012

  9. [16]

    C. I. Byrnes and A. Isidori, ”New results and examples in nonlinear feedback stabilization”, Systems and Control Letters, vol. 12, no. 5, pp. 437-442, 1989

    1989

  10. [17]

    W. P. M. H. Heemels, J. H. Sandee and P. P. J. Van den Bosch, ”Analysis of event-driven controllers for linear systems”, Internaional Journal of Control , vol. 81, no. 4, pp. 571-590, 2008

    2008

  11. [18]

    Anta and P

    A. Anta and P. Tabuada, ”To Sample or not to Sample: Self-Triggered Control for Nonlinear Systems”, IEEE Trans. Autom. Control , vol. 55, no. 9, pp. 2030-2042, 2010

    2010

  12. [20]

    Zhang, X

    G. Zhang, X. Wang, Z. Wang, and N. Pang, ”Optimized backstepping tracking control using reinforcement learning for strict-feedback nonlin- ear systems with monotone tube performance boundaries,”International Journal of Control , pp. 1–13, 2025

    2025

  13. [21]

    K. Zhou, Z. Wang, Z. Chen, and X. Wang, ”Event-triggered observer- based fixed-time consensus control for uncertain nonlinear multiagent systems with unknown states,” in 2024 International Conference on Neuromorphic Computing (ICNC) , pp. 1-7, 2024

    2024

  14. [22]

    L. Xing, C. Wen, Z. Liu, H. Su and J. Cai, ”Event-Triggered Adaptive Control for a Class of Uncertain Nonlinear Systems”, IEEE Trans. Autom. Control, vol. 62, no. 4, pp. 2071-2076, 2017

    2017

  15. [23]

    Z. Wang, H. Wang, X. Wang, N. Pang and Q. Shi, ”Event-Triggered Adaptive Neural Control for Full State-Constrained Nonlinear Systems with Unknown Disturbances” Cogn. Comput., vol. 16, no. 2, pp. 717– 726, 2023

    2023

  16. [24]

    L. Wang, X. Wang, and Z. Wang, ”Event-triggered optimal tracking control for strict-feedback nonlinear systems with non-affine nonlinear faults,” Nonlinear Dyn., vol. 112, no. 17, pp. 15413-15426, 2024

    2024

  17. [25]

    L. X. Wang and J. M. Mendel, ”Fuzzy basis functions, universal approximation, and orthogonal least-squares learning”, IEEE Trans. Neural Netw., vol. 3, no. 5, pp. 807-814, 1992

    1992

  18. [26]

    Observer-based event-triggered adaptive control for nonlinear multiagent systems with unknown states and disturbances,

    N. Pang, X. Wang, and Z. M. Wang, “Observer-based event-triggered adaptive control for nonlinear multiagent systems with unknown states and disturbances,” IEEE Trans. Neural Netw. Learn. Syst. , vol. 34, no. 9, pp. 6663–6669, 2023

    2023

  19. [27]

    Fixed-time adaptive neural tracking control for a class of uncertain nonstrict nonlinear systems,

    D. S. Ba, Y . X. Li, and S. C. Tong, “Fixed-time adaptive neural tracking control for a class of uncertain nonstrict nonlinear systems,” Neurocomputing, vol. 363, pp. 273–280, 2019

    2019

  20. [28]

    Y . Ji, X. Yin, and S. Lafortune, ”Local mean payoff supervisory control for discrete event systems,” IEEE Transactions on Automatic Control , vol. 67, no. 5, pp. 2282-2297, 2021

    2021

  21. [29]

    P. Lv, Z. Xu, Y . Ji, S. Li, and X. Yin, ”Optimal supervisory control of discrete event systems for cyclic tasks,” Automatica, vol. 164, p. 111634, 2024

    2024

Pith tools

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