Robustness in Experiment Design
The DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index.
Abstract
This paper focuses on the problem of robust experiment design, i.e., how to design an input signal which gives relatively good estimation performance over a large number of systems and model structures. Specifically, we formulate the robust experiment design problem utilizing fundamental limitations on the variance of estimated parametric models as constraints. Using this formulation we design an input signal for situations where only diffuse a priori information is known about the system. Furthermore, we present a robust version of the unprejudiced optimal input design problem. To achieve this, we first develop a closed form solution for the input spectrum which minimizes the maximum weighted integral of the variance of the frequency response estimate over all model structures.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“To ensure consistency of approximations and obtain unique solutions in each iteration step, some exploration noise must be added to control input to meet the requirements of the persistent excitation condition [40-42].”
From Data-driven control of hydraulic servo actuator: An event-triggered adaptive dynamic programming approach (2023) · cites this paper as [42]
Citations
Cites
System Identification: Theory for the User
Lennart Ljung, A. (1987) · Cited by 17,690
Rockafellar, R. T. (1970) · Princeton University Press eBooks · doi:10.1515/9781400873173 · Cited by 10,801
Garreau, G. A., & Rudin, W. (1987) · The Statistician · doi:10.2307/2348852 · Cited by 10,622
System identification—Theory for the user
Sinha, N. K. (1989) · Automatica · doi:10.1016/0005-1098(89)90019-8 · Cited by 9,184
Optimization by Vector Space Methods
(1970) · Students Quarterly Journal · doi:10.1049/sqj.1970.0088 · Cited by 6,004 · cited 1×
“This problem can now be solved using standard tools from calculus of variations [46].”
Network analysis and feedback amplifier design
Bode, H. W. (1945) · Virtual Defense Library (Ministerio de Defensa) · Cited by 3,119
de Gooijer, J. G. (1990) · International Journal of Forecasting · doi:10.1016/0169-2070(90)90013-2 · Cited by 2,245
System Identification: A Frequency Domain Approach
Pintelon, R., & Schoukens, J. (2003) · Cited by 2,157
Fedorov, V. V., Studden, W. J., & Klimko, E. M. (1972) · Cited by 2,092
Theory of Optimal Experiments.
Herzberg, A. M., Wynn, H. P., Fedorov, V. V., Studden, W. J., & Klimko, E. M. (1972) · Biometrika · doi:10.2307/2334826 · Cited by 1,897 · cited 1×
“Optimal experiment design has been studied both in the statistics literature [7]–[9] and in the engineering literature [3], [10]–[14], primarily focusing on the goals of system identification.”
System Identification: A Frequency Domain Approach
Pintelon, R., & Schoukens, J. (2012) · doi:10.1002/9781118287422 · Cited by 1,833 · cited 1×
“Basic prior knowledge about the plant can be obtained, for example, by using nonparametric frequency domain methods based on a simple experiment [4]–[6], [25], [26]; however, the use of this kind of prior knowledge for robust experiment design is not considered in the present contribution, and it will be explored in a…”
Axiomatic derivation of the principle of maximum entropy and the principle of minimum cross-entropy
Shore, J., & Johnson, R. (1980) · IEEE Transactions on Information Theory · doi:10.1109/tit.1980.1056144 · Cited by 1,774 · cited 1×
“This has already been discussed in [21], where it is noted that the results of this section resemble the development of the Principle of Maximum Entropy as given in [41] and [42], where the same issue regarding the incorporation of prior knowledge arises.”
Goodwin, G. C., Graebe, S. F., & Salgado, M. E. (2000) · Prentice Hall PTR eBooks · Cited by 1,465
Self-citationDynamic System Identification - Experiment Design and Data Analysis
(1977) · Mathematics in Science and Engineering · doi:10.1016/s0076-5392(08)x6188-0 · Cited by 1,398
Real and complex analisis / Walter Rudin
Rudin, W. (1987) · McGraw-Hill, Inc. eBooks · Cited by 1,117
System Identification Parameter and State Estimation
Eykhoff, P. (1974) · Virtual Defense Library (Ministerio de Defensa) · Cited by 1,083
(1958) · Medical Entomology and Zoology · Cited by 1,059
Functions of One Complex Variable I
Conway, J. B. (1978) · Graduate Texts in Mathematics · doi:10.1007/978-1-4612-6313-5 · Cited by 905
Achieser, N. I., & Hyman, C. J. (1956) · CERN Document Server (European Organization for Nuclear Research) · Cited by 792
Fundamental Limitations in Filtering and Control
Seron, M. M., Braslavsky, J. H., & Goodwin, G. C. (1997) · Communications and Control Engineering · doi:10.1007/978-1-4471-0965-5 · Cited by 664 · cited 2×
“As explained in detail in [24], Theorem 1 shows that it is not possible to reduce the variance of uniformly at all frequencies by choosing a suitable model structure, since if we reduce the variance at some frequencies, it will necessarily increase at others, thus implying a “water-bed” effect [30].”
Self-citationOptimal input signals for parameter estimation in dynamic systems--Survey and new results
Mehra, R. (1974) · IEEE Transactions on Automatic Control · doi:10.1109/tac.1974.1100701 · Cited by 557 · cited 1×
“Optimal experiment design has been studied both in the statistics literature [7]–[9] and in the engineering literature [3], [10]–[14], primarily focusing on the goals of system identification.”
From experiment design to closed-loop control
Hjalmarsson, H. (2005) · Automatica · doi:10.1016/j.automatica.2004.11.021 · Cited by 506 · cited 1×
“This is paradoxical since the “optimal experiment” then depends on the very thing that the experiment is aimed at estimating [13].”
Mathematical methods and theory in games, programming and economics
(1960) · Journal of the Franklin Institute · doi:10.1016/0016-0032(60)90869-3 · Cited by 411
Optimal experiment designs with respect to the intended model application
Gevers, M., & Ljung, L. (1986) · Automatica · doi:10.1016/0005-1098(86)90064-6 · Cited by 274
Least costly identification experiment for control
Bombois, X., Scorletti, G., Gevers, M., Van den Hof, P. M. J., & Hildebrand, R. (2006) · Automatica · doi:10.1016/j.automatica.2006.05.016 · Cited by 237 · cited 2×
“Work in this area has been growing in recent years [15]–[22].”
Asymptotic variance expressions for identified black-box transfer function models
Ljung, L. (1985) · IEEE Transactions on Automatic Control · doi:10.1109/tac.1985.1104093 · Cited by 236 · cited 4×
“Note that the asymptotic (in model order) variance expression [31] used to develop this equation for the input spectrum does not consider the effect of bias.”
InfluentialRobust optimal experiment design for system identification
Rojas, C. R., Welsh, J. S., Goodwin, G. C., & Feuer, A. (2007) · Automatica · doi:10.1016/j.automatica.2006.12.013 · Cited by 209
Self-citationInput design via LMIs admitting frequency-wise model specifications in confidence regions
Jansson, H., & Hjalmarsson, H. (2005) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2005.856652 · Cited by 200 · cited 1×
“Solutions for models having a finite number of parameters can be obtained, in general, only numerically, by using convex optimisation techniques [43].”
BackgroundVariational Principle for Many-Fermion Systems
Lieb, E. H. (1981) · Physical Review Letters · doi:10.1103/physrevlett.46.457 · Cited by 192
Skilling, J. (1988) · Maximum-Entropy and Bayesian Methods in Science and Engineering · doi:10.1007/978-94-009-3049-0_8 · Cited by 191 · cited 1×
“This has already been discussed in [21], where it is noted that the results of this section resemble the development of the Principle of Maximum Entropy as given in [41] and [42], where the same issue regarding the incorporation of prior knowledge arises.”
Optimal Experiment Design for Dynamic System Identification
(1979) · Lecture Notes in Control and Information Sciences · doi:10.1007/bfb0040988 · Cited by 182
Optimal experiment design for dynamic system identification
(1981) · Mathematics and Computers in Simulation · doi:10.1016/0378-4754(81)90104-x · Cited by 124
Guaranteed non-asymptotic confidence regions in system identification
Campi, M. C., & Weyer, E. (2005) · Automatica · doi:10.1016/j.automatica.2005.05.005 · Cited by 105 · cited 1×
“The possibility of removing this last condition and keeping at the same time the simplicity of our expressions, which have a very intuitive interpretation, is beyond the scope of this paper, since they are much more difficult to obtain, and most approaches to finite sample analysis are done using numerical techniques…”
Variational Principle for Many-Fermion Systems
Lieb, É. H. (2006) · Springer eBooks · doi:10.1007/3-540-27056-6_19 · Cited by 105 · cited 1×
“Proof: See the Lemma on page 458 of [47].”
The Design and Analysis of Experiments.
Rider, P. R., & Kempthorne, O. (1952) · American Mathematical Monthly · doi:10.2307/2306794 · Cited by 72
Choosing Between Open- and Closed-Loop Experiments in Linear System Identification
Agüero, J. C., & Goodwin, G. C. (2007) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2007.902756 · Cited by 49
Self-citationRobust identification of process models from plant data
Goodwin, G. C., Agüero, J. C., Welsh, J. S., Yuz, J. I., Adams, G., & Rojas, C. R. (2008) · Journal of Process Control · doi:10.1016/j.jprocont.2008.06.004 · Cited by 47
Self-citationUnprejudiced optimal open loop input design for identification of transfer functions
Yuan, Z.-D., & Lennart Ljung, A. (1985) · Automatica · doi:10.1016/0005-1098(85)90043-3 · Cited by 42
On the equivalence of least costly and traditional experiment design for control
Rojas, C. R., Agüero, J. C., Welsh, J. S., & Goodwin, G. C. (2008) · Automatica · doi:10.1016/j.automatica.2008.03.023 · Cited by 38
Self-citationFinite sample properties of indirect nonparametric closed-loop identification
Welsh, J. S., & Goodwin, G. C. (2002) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2002.800768 · Cited by 33 · cited 1×
“Basic prior knowledge about the plant can be obtained, for example, by using nonparametric frequency domain methods based on a simple experiment [4]–[6], [25], [26]; however, the use of this kind of prior knowledge for robust experiment design is not considered in the present contribution, and it will be explored in a…”
Self-citationGeometric analysis of stochastic model errors in system identification
Mårtensson, J. (2007) · KTH Publication Database DiVA (KTH Royal Institute of Technology) · Cited by 33
ROBUST INPUT DESIGN USING SUM OF SQUARES CONSTRAINTS
Mårtensson, J., & Hjalmarsson, H. (2006) · IFAC Proceedings Volumes · doi:10.3182/20060329-3-au-2901.00219 · Cited by 28
ON SOME ROBUSTNESS ISSUES IN INPUT DESIGN
Hjalmarsson, H., Mårtensson, J., & Wahlberg, B. (2006) · IFAC Proceedings Volumes · doi:10.3182/20060329-3-au-2901.00078 · Cited by 27
UTILIZING PRIOR KNOWLEDGE IN ROBUST OPTIMAL EXPERIMENT DESIGN
Goodwin, G. C., Welsh, J. S., Feuer, A., & Derpich, M. S. (2006) · IFAC Proceedings Volumes · doi:10.3182/20060329-3-au-2901.00220 · Cited by 19
Self-citationHow to Make Bias and Variance Errors Insensitive to System and Model Complexity in Identification
Mårtensson, J., & Hjalmarsson, H. (2010) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2010.2052294 · Cited by 16 · cited 1×
“Based on geometric properties of the information matrix, a modification of the least-costly identification experiment approach is analyzed in [19] and [20].”
Fundamental Limitations on the Variance of Estimated Parametric Models
Rojas, C. R., Welsh, J. S., & Agüero, J. C. (2009) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2008.2010981 · Cited by 15
Self-citationRelative Error Issues in Sampled Data Models
Goodwin, G. C., Yuz, J. I., & Agüero, J. C. (2008) · IFAC Proceedings Volumes · doi:10.3182/20080706-5-kr-1001.00848 · Cited by 14
Self-citationGood, Bad and Optimal Experiments for Identification
Goodwin, G. C., Rojas, C. R., & Welsh, J. S. (2006) · Cited by 13
Self-citationZhou, T. (2006) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2005.860279 · Cited by 10 · cited 1×
“Basic prior knowledge about the plant can be obtained, for example, by using nonparametric frequency domain methods based on a simple experiment [4]–[6], [25], [26]; however, the use of this kind of prior knowledge for robust experiment design is not considered in the present contribution, and it will be explored in a…”
Evaluation and comparison of robust optimal experiment design criteria
Welsh, J. S., Goodwin, G. C., & Arie Feuer, B. (2006) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.1109/acc.2006.1656457 · Cited by 9 · cited 1×
“As argued in [16], [32]–[34], absolute variances are not particularly useful when one wants to design an experiment that applies to a broad class of systems.”
Self-citationSampled Data Errors-in-Variables Systems
Agüero, J. C., Goodwin, G. C., Söderström, T., & Yuz, J. I. (2009) · IFAC Proceedings Volumes · doi:10.3182/20090706-3-fr-2004.00192 · Cited by 8 · cited 1×
“However, it is a standard assumption for the derivation of several variance results, e.g., see [31].”
BackgroundSelf-citationRojas, C. R., Agüero, J.-C., Welsh, J. S., Goodwin, G. C., & Feuer, A. (2008) · Cited by 8
Self-citationOptimal experiment design with diffuse prior information
Rojas, C. R., Goodwin, G. C., Welsh, J. S., & Feuer, A. (2007) · European Control Conference (ECC) · doi:10.23919/ecc.2007.7068400 · Cited by 4 · cited 9×
“The following lemma, from [21], describes how must be chosen to satisfy Criterion B.”
InfluentialSelf-citation
Research works printed but not linked: 3
- goodwin, 2006, Utilizing prior knowledge in robust optimal experiment design, Proc 14th IFAC Symp System Identification, 1358
- duren, 1970, Theory of Spaces
- koosis, 1998, Introduction to Spaces
Cited by
Fault-tolerant control of a hydraulic servo actuator via adaptive dynamic programming
Stojanović, V. (2023) · Mathematical Modelling and Control · doi:10.3934/mmc.2023016 · Cited by 142
Adaptive Input Design for Identification of Output Error Model with Constrained Output
Stojanović, V., & Filipović, V. (2013) · Circuits Systems and Signal Processing · doi:10.1007/s00034-013-9633-0 · Cited by 111
Djordjevic, V., Tao, H., Song, X., He, S., Gao, W., & Stojanovic, V. (2023) · Mathematical Biosciences and Engineering · doi:10.3934/mbe.2023376 · Cited by 107
Stojanović, V., Nedić, N. N., Pršić, D., & Dubonjić, L. (2016) · Applied Mathematical Modelling · doi:10.1016/j.apm.2016.02.014 · Cited by 110
Data-driven control of hydraulic servo actuator based on adaptive dynamic programming
Djordjevic, V., Stojanović, V., Tao, H., Song, X., He, S., & Gao, W. (2022) · Discrete and Continuous Dynamical Systems - S · doi:10.3934/dcdss.2021145 · Cited by 98
Metadata
- Authors
- Rojas, C. R., Aguero, J.-C., Welsh, J. S., Goodwin, G. C., & Feuer, A.
- Affiliations
- KTH Royal Institute of Technology (SE)
- University of Newcastle Australia (AU)
- Venue
- IEEE Transactions on Automatic Control
- Published
- 2012-04
- Volume
- 57
- Issue
- 4
- Pages
- 860-874
- ISSN
- 0018-9286 (ISSN-L)
- 1558-2523
- OpenAlex
- W2137921991
- Cited by
- 27
- References
- 53
- Citations by year
- 2024: 2 · 2023: 3 · 2022: 1 · 2021: 3 · 2019: 2 · 2018: 2 · 2017: 3 · 2016: 2 · 2015: 2 · 2014: 2 · 2013: 3 · 2012: 2
- FWCI
- 3.57
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Transactions on Automatic Control (publishedVersion)
- NOVA (University of Newcastle Australia) (submittedVersion)
- Topics
- Physical Sciences › Engineering › Control and Systems Engineering › Control Systems and Identification
- Social Sciences › Decision Sciences › Management Science and Operations Research › Optimal Experimental Design Methods
- Physical Sciences › Mathematics › Applied Mathematics › Statistical and numerical algorithms
- Keywords
- input design, frequency response estimation, model structure uncertainty
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.