Systems and Control · arXiv:2611.17046 v1 · submitted 12 Nov 2026 · 18 references
Robust Control Design for LTI Systems with Sampling-Time Jitter
The paper says
“Sampling-time jitter makes a sampled-data control loop time-varying, so standard LTI robust-control synthesis does not apply directly. This paper reformulates dynamic output feedback design under bounded, unmeasured sample-to-sample jitter as a robust-control problem.”
more of the abstract
The exact zero-order-hold model is decomposed into a nominal time-invariant plant and a memoryless, time-varying uncertainty block with a norm bound representing the sampling-time jitter. As the block is memoryless, the operator small-gain condition for the time varying loop reduces to a standard robustness test with a matrix norm bound. It is shown that the same test also guarantees uniform exponential stability for every admissible jitter sequence. Embedding the decomposition into a generalized plant with performance channels serves two synthesis objectives with standard robust-control tools: Robust stability with optimized nominal performance through structured H-inf synthesis, and robust performance through mu synthesis with constant scalings. A numerical example shows that jitter can destabilize a control loop with a nominal H-inf controller, whereas both robust designs remain stable and the robust-performance design is less sensitive to jitter at the cost of nominal performance.
from the abstract
Pith says
Machine review · deepseek-v4-flash · 13 Nov 2026 · what was read was not recorded · not checked by a person
It can be wrong. Challenge this review or reply to it.
Holds up, with conditions. Solid small-gain packaging of jitter as a full-block uncertainty; the theory is clean and the workflow is practical. The only real gap is that the numerical example's certificate relies on a grid-computed δ̄ that the paper's own rigorous bound would invalidate. This line is wrong
3 major objections · 8 minor
How often objections from reviews like this one are right: not yet measured
Not checked: figures, code, what the 18 references say.
The paper says, at §VI, Appendix
The review did not quote this passage.
A classifier read this sentence of the paper as acknowledging the concern (probability 0.81):
“A finite grid can miss the true maximum between grid points, so its ¯δ can sit slightly below the true one.”
Pith says: Objection 1 of 3 · major
The numerical example's certificate relies on δ̄ = 0.02832 computed from a 1000-point grid (b143). The authors acknowledge that a finite grid can miss the true maximum and mention a Lipschitz bound or the Appendix bound as remedies, but neither is actually applied.
more of objection 1
The Appendix bound (23)–(24) yields δ̄ ≈ 0.0318, which is about 12% larger. Using this rigorous bound, the small-gain product for Krob would become 0.90 × (0.0318/0.02832) ≈ 1.01, violating condition (18). Thus the example does not demonstrate a valid certificate for the reported controllers. The authors should either (a) recompute δ̄ rigorously (e.g., via the Lipschitz argument they mention or the Appendix bound) and verify that the small-gain condition still holds, or (b) adjust the design parameters (ε̄, weights) so that the certificate is satisfied with a rigorous δ̄. Without this, the central claim of 'provably correct controllers' is not supported by the example.
The paper says, at §V-B, Eq. (20)–(21)
The review did not quote this passage.
Pith says: Objection 2 of 3 · major
The robust-performance synthesis uses constant scalings (FitOrder=0) and the authors correctly note that D–K iteration need not converge to the global optimum (b131). However, the certificate check in Step 3 (Tab. I) requires that the fitted bound be below one for the selected controller.
more of objection 2
In the example, the reported fitted bound is 0.8365 < 1, but this is the value from the musyn iterate, not necessarily a rigorous upper bound on the scaled H∞ norm. The authors should clarify whether this value is the final PeakMuFit or a post-iteration check with the constant scaling fixed. More importantly, they should state whether the condition (20) is verified a posteriori with the chosen constant scalings, or whether the reported number is simply the musyn output. If the latter, the certificate is only as reliable as the D–K iteration's termination.
The paper says, at §IV-C, Theorem 2
The review did not quote this passage.
Pith says: Objection 3 of 3 · major
The proof of Theorem 2 uses the discrete-time bounded-real lemma to obtain a common quadratic Lyapunov function. The argument assumes that the normalized return map Ñ₁₁ = δ̄N₁₁ has H∞ norm < 1 and that the uncertainty satisfies ‖ẽ_Δ(ε_k)‖₂ ≤ 1.
more of objection 3
However, the bounded-real lemma requires that the realization (A_N, δ̄B_N, C_N, δ̄D_N) be minimal or at least detectable and stabilizable. The authors should comment on whether the nominal closed-loop realization is minimal (or at least that the uncontrollable/unobservable modes are stable, which follows from internal stability). This is a minor technical point but should be addressed for completeness.
8 minor objections
The paper says, at §III-D, Eq. (12)
Pith says: minor objection 1 of 8
The notation ‖D‖_{ℓ₂→ℓ₂} = sup_k ‖Δ(ε_k)‖₂ is correct for a memoryless operator, but the equality to max_{|ε|≤ε̄} σ̄(Δ(ε)) holds only if the supremum over k is attained at some ε in the closure of the range. Since the range is compact and the singular value is continuous, this is fine. Consider adding a brief justification.
The paper says, at §IV-A, Eq. (14)
Pith says: minor objection 2 of 8
The generalized plant P includes y_Δ,k as an input and u_Δ,k as an output. The equation for x_{k+1} includes + y_Δ,k, but the uncertainty output is y_Δ,k = Δ(ε_k) u_Δ,k. This is consistent with Fig. 3. However, the text says 'The inputs of P are [y_Δ, r, u] and its outputs are [u_Δ, z, v].' This is slightly confusing because u is the control input and v is the controller input. Clarify that the ordering matches the standard P–K–Δ convention.
show 6 more
The paper says, at §V-C
Pith says: minor objection 3 of 8
The design-jitter-bound discussion (b135–b138) is important but somewhat buried. Consider moving it to a separate subsection or highlighting it in the workflow figure.
The paper says, at §VI, b143–b145
Pith says: minor objection 4 of 8
The sentence 'Both raise δ̄ and are omitted due to space limitation' is unsatisfactory for a journal paper. Either include the rigorous bound or remove the claim that the example certificates are rigorous. This is related to major comment 1.
The paper says, at §VI, Fig. 7
Pith says: minor objection 5 of 8
The alternating jitter sequence ε_k = 0.15·(-1)^k is a specific periodic pattern. The authors correctly note that this is not a claimed worst case (b172). However, the caption could be more precise: 'Alternating jitter, ε_k = ±0.15' is clear, but the text should also state that the nominal responses are shown for reference.
The paper says, at §II
Pith says: minor objection 6 of 8
The related work section is concise but could mention the input-delay approach (Fridman et al.) more explicitly in the context of output-feedback synthesis. Currently it only cites [8] without discussing its limitations relative to the proposed method.
The paper says, at General
Pith says: minor objection 7 of 8
The paper uses the notation ‖·‖_{H∞} for the H∞ norm of a discrete-time system. This is standard but could be clarified as the ℓ₂-induced norm of the LTI operator.
The paper says, at Appendix, Eq. (23)–(24)
Pith says: minor objection 8 of 8
The bound for δ̄_B uses the integral from t_s to (1+ε̄)t_s. For negative jitter (ε_k < 0), the interval is shorter, so the tail length is |ε_k| t_s, but the majorant e^{‖A‖₂ τ} is increasing, so the bound still holds.
more of objection 8
This is correctly noted. However, the derivation of δ̄_A uses submultiplicativity: ‖e^{εA t_s} - I‖₂ ≤ e^{|ε|‖A‖₂ t_s} - 1. This is valid for any matrix norm, but the 2-norm is not submultiplicative for the exponential series? Actually, the matrix exponential satisfies ‖e^M‖ ≤ e^{‖M‖} for any submultiplicative norm, and the 2-norm is submultiplicative. So it's fine. Consider adding a brief justification.
What would settle it
Recompute the example's certificates with a proven δ̄: (1) evaluate the Appendix closed-form bound (23)–(24) for the example plant (A = [0 1; −1 −0.2], B = [0;1], t_s = 0.1, ε̄ = 0.2) and compare δ̄_appendix against the grid value 0.02832; (2) with δ̄_appendix re-evaluate Krob's product δ̄‖N_11‖_H∞ and Kµ's scaled bound (20) using the reported controllers and fitted scalings. If the Krob product exceeds 1 (I estimate ≈1.01) or the Kµ bound exceeds 1, the Section VI certificate claims (b155–b157) do not survive the paper's own rigorous bound, and the example must be re-run with a validated δ̄ (Lipschitz-refined grid) or resynthesized controllers. A secondary check: obtain the scripts and reproduce δ̄ = 0.02832, the products 0.90 and 0.8365, and the alternating-jitter divergence of Knom, to confirm the printed numbers.
from the skeptic's reading
How the review went
4 readings by deepseek-v4-flash · 13 Nov 2026
- Reader holds up, with conditions
- Skeptic partly agrees
- Editor wrote the verdict above
- Referee major revision: 3 major objections, 8 minor
Not run on this paper: the simulated authors' defence.
Pith did not check this review's objections against the paper.
The review in full
The review in brief, machine-written
The paper does something genuinely useful: it takes the known exact ZOH jitter decomposition (Fujioka, Kao-Fujioka) and shows that because the jitter block is memoryless, the operator small-gain theorem reduces to a simple matrix-norm condition that can be checked with standard LTI tools. Theorems 1 and 2 are correctly argued and give a clean certificate for finite-gain stability and uniform exponential stability. The synthesis workflow—structured H∞ for robust stability with nominal performance, constant-scaled μ synthesis for robust performance—is a practical contribution that closes an output-feedback gap Fujioka and Nakai left open. The paper is well written, the math is internally coherent, and the limitations (conservatism of the full block, need for a known jitter envelope) are stated honestly.
The soft spot is in the numerical example. The paper computes δ̄ = 0.02832 from a 1000-point grid and acknowledges that a grid can understate the true maximum. The stress-test note points out that the only rigorous bound the paper supplies (Appendix (23)–(24)) gives δ̄ ≈ 0.0318, about 12% larger. Applied to the reported numbers, the small-gain product for Krob would go from 0.90 to about 1.01, and for Kμ from 0.8365 to about 0.94. So the example does not actually demonstrate a proof-grade certificate with the paper's own rigorous bound. The authors mention that a Lipschitz bound or the Appendix bound would close the gap but omit the calculation due to space. That is a real weakness in the demonstration, though it does not affect the theoretical validity of the theorems.
Who is this for? Practitioners who want a turnkey robust-control workflow for jitter, and theorists who want a clean application of small-gain to sampled-data systems. The paper deserves a serious referee; the theory is sound and the workflow is useful. The example needs to be fixed—either recompute with a rigorous δ̄ or show that the grid value is actually valid (e.g., by a Lipschitz argument). I would accept peer review on this paper and recommend conditional acceptance with a request to strengthen the numerical demonstration.
The strongest objection
The central theorems are internally sound. Theorem 1 is a correct application of operator small-gain to the exact ZOH jitter decomposition: Δ(ε_k) is memoryless, (12) gives a uniform gain bound, and (18) closes the loop uniformly over E(ε̄). Theorem 2's bounded-real-lemma step is valid: (A_N, δ̄B_N, C_N, δ̄D_N) realizes Ñ_11 = δ̄N_11 with ||Ñ_11||∞ < 1, the BRL supply term is made nonpositive by ||ẽ_Δ(ε_k)|| ≤ 1, and the construction (14) makes N_11 strictly proper (the perturbation enters only the state update), so well-posedness is immediate. The constant-scaling robust-performance argument (20)–(21) is also correct: the scalar port commutation D_i bΔ = bΔ D_o preserves the unit gain of the time-varying block, and dynamic scalings are correctly excluded (b126–b129).
The load-bearing concern is at the certificate-verification step of the demonstration. The example computes δ̄ = 0.02832 by a 1000-point grid; the authors acknowledge (b143) a grid can understate the true maximum, deferring rigor to a Lipschitz check or the Appendix bound (b145). The grid error itself is negligible: σ̄(Δ(ε)) has Lipschitz constant ≈0.16 for this plant, so 1000 points over [−0.2, 0.2] give error ≲3e-5. The material gap is between the grid value and the only rigorous bound the paper supplies. For A = [0 1; −1 −0.2], B = [0;1], t_s = 0.1, ε̄ = 0.2, (23)–(24) give ‖A‖₂ ≈ 1.105, δ̄_A ≈ 0.0223, δ̄_B ≈ 0.0226, hence δ̄_appendix ≈ 0.0318 — about 12% above the grid value. Applied to the reported numbers, Krob's product 0.90 becomes ≈1.01 (the 10% margin is consumed and exceeded) and Kµ's 0.8365 becomes ≈0.94. The paper's only worked example therefore does not demonstrate a proof-grade certificate: with its own rigorous δ̄, the headline robust-stability controller likely fails (18), and the scripts are only 'available on request', so the numbers cannot be independently checked. The design-jitter-bound caveat (b135–b138) is acknowledged scoping, not part of this concern.
I read the paper's claims as: (i) the jittered loop is exactly an LFT of a nominal LTI system with a memoryless norm-bounded perturbation; (ii) the small-gain condition (18) certifies finite-gain stability and UES uniformly over E(ε̄); (iii) standard tools (systune, musyn with FitOrder=0) can synthesize controllers satisfying the certificate; (iv) the example demonstrates the workflow. Claims (i)–(iii) are well supported. I checked Theorem 2's bounded-real-lemma step in detail — the normalized realization, the strict norm bound, the pointwise supply-rate argument, and well-posedness all go through — and the constant-scaling robust-performance algebra commutes correctly for the time-varying block, with the exclusion of dynamic scalings correct for the time-varying case. The reader's low correctness risk is accurate for the theorems.
The weakest point is the demonstration of (iv). A proof-grade certificate requires a proven δ̄, and the example uses a grid value that is not proven; the authors acknowledge this (b143), but the only rigorous bound they offer (Appendix (23)–(24)) is so much larger that the example's margins vanish. My estimate (δ̄_appendix ≈ 0.0318 vs 0.02832) puts Krob at ≈1.01, meaning the paper's own rigorous bound likely rejects the controller it presents as certified, while Kµ survives with a thinner margin. The grid-error contribution itself is tiny (≈3e-5, since σ̄(Δ(ε)) is Lipschitz with constant ≈0.16), so this is not a failure of the grid as a numerical tool; it is the absence of a demonstrated proven δ̄, which is precisely the guarantee the paper advertises. This is fixable (Lipschitz-refined bound, shipped code, or a resynthesized example) and does not touch Theorems 1–2. The reader's CONDITIONAL verdict already rests on the validity of the jitter envelope and the example certificates; my analysis sharpens the second condition without changing the verdict, so I recommend UNCHANGED.
The referee's report
If the results hold, the paper provides a practical bridge between aperiodic sampling theory and off-the-shelf robust-control software, enabling engineers to certify jitter robustness without dedicated LMI solvers or hybrid-system tools. The key insight—that the memoryless nature of the jitter perturbation reduces the operator small-gain condition to a matrix-norm test—is elegant and potentially impactful for embedded control applications.
Pith's reading of the paper (a machine's summary, not the authors')
This paper shows that standard off-the-shelf robust-control tools can design controllers for sampled-data systems whose sampling intervals vary unpredictably from sample to sample. The authors decompose the jittered plant into a nominal time-invariant part and a memoryless, norm-bounded uncertainty block. Because the uncertainty block has no internal dynamics, the operator small-gain condition for the time-varying loop reduces to a simple matrix-norm test that can be checked with a single H-infinity norm. The same test also guarantees uniform exponential stability of the closed-loop state, including behavior between samples. The paper then embeds this decomposition into a generalized plant and provides two synthesis workflows: structured H-infinity synthesis for robust stability with optimized nominal performance, and mu synthesis with constant scalings for guaranteed robust performance. A numerical example with a lightly damped second-order plant demonstrates that a nominal controller stable at every constant jitter value can destabilize when jitter alternates, whereas the robust designs remain stable. The main practical result is that an engineer can certify jitter robustness using standard robust-control toolbox commands, provided a known bound on the jitter is available.
What carries the argument. The key machinery is the decomposition of the jittered exact ZOH model into a nominal LTI part and a memoryless, norm-bounded uncertainty block. The jitter block is defined as the full matrix [Delta_A(epsilon_k) | Delta_B(epsilon_k)] that multiplies the stacked state-input vector. Because this block depends only on the current jitter value and has no internal state, its induced $\ell^2$ gain is simply the maximum singular value over the jitter envelope. This static property is what reduces the operator small-gain condition to the matrix-norm test (18), and it also enables the use of constant scalings in mu synthesis without breaking the time-varying certificate.
Core claim. The central discovery is that sampling-time jitter in a sampled-data control loop can be exactly represented as a memoryless, norm-bounded uncertainty block in feedback with a nominal LTI plant. Because the uncertainty block has no dynamics, the operator small-gain theorem for time-varying systems reduces to a static matrix-norm condition: the product of the norm bound of the jitter block and the H-infinity norm of the nominal return map must be less than one. This condition simultaneously guarantees well-posedness, finite-gain stability, and uniform exponential stability for every admissible jitter sequence. The paper further shows that this decomposition can be embedded into a generalized plant with performance channels, allowing standard robust-control synthesis tools (structured H-infinity synthesis and constant-scaled mu synthesis) to produce controllers with provable jitter-robustness certificates.
The premise it rests on. The certificates require a known, never-violated bound on the relative jitter for every sample, and the engineer must guarantee that the actual jitter never exceeds this bound.
If the paper is right (Pith's inferences, not the authors')
Engineers can now certify jitter robustness of a control loop using standard H-infinity and mu-synthesis tools, without needing dedicated LMI solvers or hybrid-system analysis.
The small-gain certificate (18) provides a simple, LTI-computable test that can be applied a posteriori to any existing discrete-time controller to check its robustness against a given jitter envelope.
The design-jitter bound can be used as a tuning knob: a smaller certified bound allows better nominal performance, while a larger bound gives stronger jitter guarantees at the cost of performance.
The certificate covers not only input-output stability but also uniform exponential stability of the state, including inter-sample behavior, without requiring an explicit inter-sample performance specification.
The mu-synthesis approach with constant scalings provides a guaranteed worst-case induced gain from reference to weighted error, valid for every admissible jitter sequence.
The paper's approach could be extended to handle combined jitter and plant uncertainty by building a joint uncertainty model with a structured block, though the authors note this requires a new certificate.
The constant-scaling requirement for mu synthesis means that frequency-dependent D-scales cannot be used, which is a significant restriction compared to standard mu synthesis for LTI uncertainty. This limits the achievable robust performance bound.
The numerical example suggests that the certificates are moderately conservative: the small-gain test certifies stability up to epsilon_bar=0.22 for the robust-stability controller, while a search over periodic jitter sequences finds instability only at epsilon=0.415. This gap indicates room for less conservative analysis methods.
The paper's framework assumes the jitter bound is known and never violated. In practice, measuring or guaranteeing such a bound from a real-time operating system is nontrivial, and the engineer takes on responsibility for sequences outside the chosen envelope.
Check this review
Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.
How often reviews like this one are right
Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.
How often objections from reviews like this one are right: not yet measured
Reviews by people
This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.
The paper
Abstract
Sampling-time jitter makes a sampled-data control loop time-varying, so standard LTI robust-control synthesis does not apply directly. This paper reformulates dynamic output feedback design under bounded, unmeasured sample-to-sample jitter as a robust-control problem. The exact zero-order-hold model is decomposed into a nominal time-invariant plant and a memoryless, time-varying uncertainty block with a norm bound representing the sampling-time jitter. As the block is memoryless, the operator small-gain condition for the time varying loop reduces to a standard robustness test with a matrix norm bound. It is shown that the same test also guarantees uniform exponential stability for every admissible jitter sequence. Embedding the decomposition into a generalized plant with performance channels serves two synthesis objectives with standard robust-control tools: Robust stability with optimized nominal performance through structured H-inf synthesis, and robust performance through mu synthesis with constant scalings. A numerical example shows that jitter can destabilize a control loop with a nominal H-inf controller, whereas both robust designs remain stable and the robust-performance design is less sensitive to jitter at the cost of nominal performance.
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.
abstract compiled · 2026-11-13
The authors' main claim, as the classifier picked it out 0.45
“A numerical example shows that jitter can destabilize a control loop with a nominal H-inf controller, whereas both robust designs remain stable and the robust-performance design is less sensitive to jitter at the cost of nominal performance.”
How the main claim is stated: asserted 1.00
What the abstract itself states, as probabilities: claims first 0.03 · states limitation 0.57 · human participants 0.01 · controlled comparison 0.52 · randomized assignment 0.01 · states quantitative result 0.09 · compares with prior results 0.85 · mentions code or data availability 0.02
The abstract, sentence by sentence
This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.
| Sentence | Role | The authors' own work | States the main conclusion | Strength |
|---|---|---|---|---|
| Sampling-time jitter makes a sampled-data control loop time-varying, so standard LTI robust-control synthesis does not apply directly. | background 0.99 | 0.09 | 0.03 | asserted 1.00 |
| This paper reformulates dynamic output feedback design under bounded, unmeasured sample-to-sample jitter as a robust-control problem. | contribution 0.99 | 0.93 | 0.14 | asserted 1.00 |
| The exact zero-order-hold model is decomposed into a nominal time-invariant plant and a memoryless, time-varying uncertainty block with a norm bound representing the sampling-time jitter. | method 0.94 | 0.91 | 0.07 | asserted 1.00 |
| As the block is memoryless, the operator small-gain condition for the time varying loop reduces to a standard robustness test with a matrix norm bound. | method 0.48 | 0.81 | 0.14 | asserted 1.00 |
| It is shown that the same test also guarantees uniform exponential stability for every admissible jitter sequence. | result 0.97 | 0.91 | 0.32 | asserted 1.00 |
| Embedding the decomposition into a generalized plant with performance channels serves two synthesis objectives with standard robust-control tools: Robust stability with optimized nominal performance through structured H-inf synthesis, and robust performance through mu synthesis with constant scalings. | method 0.72 | 0.89 | 0.19 | asserted 1.00 |
| A numerical example shows that jitter can destabilize a control loop with a nominal H-inf controller, whereas both robust designs remain stable and the robust-performance design is less sensitive to jitter at the cost of nominal performance. | result 1.00 | 0.94 | 0.30 | asserted 1.00 |
What kind of paper the classifier reports this is: method or tool 0.87 · evidence from proof or derivation 0.97
What was examined, and what was not
Sources read
- arXiv listing abstract v1
Not examined
- full text
- figures and tables
- references and citing sentences
- relations to other papers
Levels run
- P abstract
Not run
- S sections
- T body sentences
- C citations
- R cross-paper relations
How the labels were made
Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-11-13: 1 request, 5,656 input tokens.
Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON
Figures (7)
Showing 6 of the paper's 7 figures. Figures are the authors', reproduced from the paper.
References (18)
- [1] D. Schwarzmann, S. K ¨aser, and J. Lunze, “Effect of sampling-time jitter on embedded control dynamics,” inProceedings of the IFAC Congress, 2026, pp. 4393–4398 on Pith
- [2] H. Fujioka, “A discrete-time approach to stability analysis of systems with aperiodic sample-and-hold devices,”IEEE Transactions on Automatic Control, vol. 54, no. 10, pp. 2440–2445, 2009 on Pith
- [3] ——, “Stability analysis for a class of networked/embedded control systems: Output feedback case,” inProceedings of the 17th IFAC World Congress, 2008 on Pith
- [4] H. Fujioka and T. Nakai, “Stabilising systems with aperiodic sample- and-hold devices: state feedback case,”IET Control Theory & Appli- cations, vol. 4, no. 2, pp. 265–272, 2010 on Pith
- [5] C.-Y . Kao and H. Fujioka, “On stability of systems with aperiodic sampling devices,”IEEE Transactions on Automatic Control, vol. 58, no. 8, pp. 2085–2090, 2013 on Pith
- [6] L. Hetel, C. Fiter, H. Omran, A. Seuret, E. Fridman, J.-P. Richard, and S.-I. Niculescu, “Recent developments on the stability of systems with aperiodic sampling: An overview,”Automatica, vol. 76, pp. 309–335, 2017 on Pith
- [7] P. Naghshtabrizi, J. P. Hespanha, and A. R. Teel, “On the robust stability and stabilization of sampled-data systems: A hybrid system approach,” inProceedings of the 45th IEEE Conference on Decision and Control (CDC), 2006, pp. 4873–4878 on Pith
- [8] E. Fridman, A. Seuret, and J.-P. Richard, “Robust sampled-data stabilization of linear systems: an input delay approach,”Automatica, vol. 40, pp. 1441–1446, 2004 on Pith
- [9] V . Suplin, E. Fridman, and U. Shaked, “Sampled-data h∞ control and filtering: Nonuniform uncertain sampling,”Automatica, vol. 43, pp. 1072–1083, 2007 on Pith
- [10] C. Briat, “Convex conditions for robust stability analysis and stabi- lization of linear aperiodic impulsive and sampled-data systems under dwell-time constraints,”Automatica, vol. 49, no. 11, pp. 3449–3457, 2013 on Pith
- [11] G. Mustafa, “Analysis and synthesis of nonuniformly sampled systems,” Ph.D. dissertation, University of Alberta, 2012 on Pith
- [12] K. Zhou, J. C. Doyle, and K. Glover,Robust and Optimal Control. Upper Saddle River, NJ: Prentice Hall, 1996 on Pith
- [13] C. A. Desoer and M. Vidyasagar,Feedback Systems: Input–Output Properties. New York: Academic Press, 1975 on Pith
- [14] A. Packard and J. Doyle, “The complex structured singular value,” Automatica, vol. 29, no. 1, pp. 71–109, 1993 on Pith
- [15] J. S. Shamma, “Robust stability with time-varying structured uncer- tainty,”IEEE Transactions on Automatic Control, vol. 39, no. 4, pp. 714–724, 1994 on Pith
- [16] S. Boyd, L. E. Ghaoui, E. Feron, and V . Balakrishnan,Linear Matrix Inequalities in System and Control Theory, ser. SIAM Studies in Applied Mathematics. Philadelphia, PA: SIAM, 1994, vol. 15 on Pith
- [17] P. Apkarian and D. Noll, “Nonsmooth h∞ synthesis,”IEEE Transac- tions on Automatic Control, vol. 51, no. 1, pp. 71–86, 2006 on Pith
- [18] P. Gahinet and P. Apkarian, “Decentralized and fixed-structure h∞ control in MATLAB,” inProceedings of the 50th IEEE Conference on Decision and Control and European Control Conference (CDC-ECC), 2011, pp. 8205–8210 on Pith
Challenges and discussion
Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.
Come back by email
A short list of the week's newly reviewed papers. Nothing is sent until you confirm from your inbox, and every email has a link that stops it. No account.
No comments yet.
Sign in to comment