System identification for complex dynamical systems: a survey

Abstract

System identification provides the data-to-model link for analysis, prediction, diagnosis, and control of dynamical systems. As engineering and scientific systems become increasingly nonlinear, high-dimensional, networked, time-varying, and partially observed, identification methods must handle noisy data, incomplete prior knowledge, safety constraints, and downstream control requirements. This review surveys system identification for complex dynamical systems through a unified framework that connects model structure, operating mode, estimation target, and computational setting. We first revisit single-system identification, including regularized regression, sparse Bayesian learning, Kalman-type filtering, and recent structured recursive methods such as auxiliary-model, multi-innovation, hierarchical, filtering-based, and coupled identification. We then review multi-system identification for families of interacting or related subsystems, covering deep neural architectures, sparse equation discovery for ordinary and partial differential equations, hybrid and topology-aware modeling, and meta-learning-based fast adaptation. Finally, we discuss large language model-assisted workflows for symbolic regression, equation discovery, feature generation, and scientific modeling. By comparing these methodological streams, this review highlights their strengths, limitations, and applicability to prediction, synchronization, diagnosis, and control. We identify key open challenges, including robust identification under biased or limited data, uncertainty-aware modeling, interpretable learning, reproducible benchmarking, and tighter integration between identification, experiment design, and control synthesis.

Keywords

System identification / complex dynamical systems / sparse Bayesian learning / recursive identification / multi-system identification / equation discovery / large language models

References

[1] Pillonetto G.,Aravkin A.,Gedon D.,Ljung L.,Ribeiro A. H.,Schon T. B.. Deep networks for system identification: a survey Automatica 2025 171 111907

[2] Wang Q. G.,Zhang L.. System identification in the network era: a survey of data issues and innovative approaches IEEE/CAA J. Autom. Sin. 2025 12 1305 19

[3] Åström K. J.,Eykhoff P.. System identification-a survey Automatica 1971 7 123 62

[4] Wang H.,Zhang Q.,Liu X.. Auxiliary model-based maximum likelihood multi-innovation recursive least squares identification for multiple-input multiple-output systems J. Frankl. Inst. 2024 361 107352

[5] Xing H.,Ding F.,Zhang X.,Luan X.,Yang E.. Highly-efficient filtered hierarchical identification algorithms for multiple-input multiple-output systems with colored noises Syst. Control Lett. 2024 186 105762

[6] Liu Q.,Chen F.,Guo Q.,Wang X.. Distributed identification based partially-coupled recursive generalized extended least squares algorithm for multivariate input-output-error systems with colored noises from observation data J. Comput. Appl. Math. 2024 449 115976

[7] Wang H.,Liu X.. Hierarchical maximum likelihood multi-innovation identification methods for a class of multivariable Hammerstein-input-nonlinear systems J. Frankl. Inst. 2025 362 108143

[8] Zhou Y.,Wu Z.,Chen Z.,Ma H.. Dynamic factor and multi-innovation-based output-input feedback Elman network modeling from measurements IEEE Trans. Instrum. Meas. 2025 74 1 13

[9] Qin H.,Ji Y.. Two-stage gradient-based iterative algorithms for the Hammerstein nonlinear systems with non-uniform sampling using the key term separation J. Frankl. Inst. 2026 363 108416

[10] Hjalmarsson H.. System identification of complex and structured systems Eur. J. Control 2009 15 275 310

[11] Tibshirani R.. Regression shrinkage and selection via the lasso J. R. Stat. Soc. Ser. B Stat. Methodol 1996 58 267 88

[12] Hoerl A. E.,Kennard R. W.. Ridge regression: biased estimation for nonorthogonal problems Technometrics 2012 12 55 67

[13] Zou H.,Hastie T.. Regularization and variable selection via the elastic net J. R. Stat. Soc. Ser. B Stat. Methodol. 2005 67 301 20

[14] Tipping M. E.. Sparse Bayesian learning and the relevance vector machine J. Mach. Learn. Res. 2001 1 211 44

[15] Wipf D.,Rao B. D.. Sparse Bayesian learning for basis selection IEEE Trans. Signal Process. 2004 52 2153 64

[16] Ji S.,Xue Y.,Carin L.. Bayesian compressive sensing IEEE Trans. Signal Process. 2008 56 2346 56

[17] Zhang Z.,Rao B. D.. Sparse signal recovery with temporally correlated source vectors using sparse Bayesian learning IEEE J. Sel. Top. Signal Process. 2011 5 912 26

[18] Zhang Z.,Rao B. D.. Extension of SBL algorithms for the recovery of block sparse signals with intra-block correlation IEEE Trans. Signal Process. 2013 61 2009 15

[19] Sant A.,Leinonen M.,Rao B. D.. Block-sparse signal recovery via general total variation regularized sparse Bayesian learning IEEE Trans. Signal Process. 2022 70 1056 71

[20] Kalman R. E.. A new approach to linear filtering and prediction problems J. Basic Eng. 1960 82 35 45

[21] Ljung L.. Asymptotic behavior of the extended Kalman filter as a parameter estimator for linear systems IEEE Trans. Automat. Contr. 1979 24 36 50

[22] Julier S. J.,Uhlmann J. K.. Unscented filtering and nonlinear estimation Proc. IEEE 2004 92 401 22

[23] Loh C. H.,Tou I. C.. A system identification approach to the detection of changes in both linear and non-linear structural parameters Earthq. Eng. Struct. Dyn. 1995 24 85 97

[24] Cui M.,Khodayar M.,Chen C.,Wang X.,Zhang Y.,Khodayar M. E.. Deep learning-based time-varying parameter identification for system-wide load modeling IEEE Trans. Smart Grid 2019 10 6102 14

[25] Simpkins A.. System identification: theory for the user, 2nd Edition (Ljung L., 1999) [On the Shelf] IEEE Robot. Automat. Mag. 2012 19 95 6

[26] Van Den Hof P. M.,Schrama R. J.. Identification and control-closed-loop issues Automatica 1995 31 1751 70

[27] Hjalmarsson H.,Gevers M.,De Bruyne F.. For model-based control design, closed-loop identification gives better performance Automatica 1995 32 1659 73

[28] Feng J.,Jia K.,Qin C.,Zhu S.,Yang X.,Tian J.. Sparse Bayesian reconstruction method for multispectral bioluminescence tomography Chin. Opt. Lett. 2010 8 1010 14

[29] Donoho D. L.. Compressed sensing IEEE Trans. Inform. Theory 2006 52 1289 306

[30] Tropp J. A.. Greed is good: algorithmic results for sparse approximation IEEE Trans. Inform. Theory 2004 50 2231 42

[31] Tropp J. A.,Gilbert A. C.. Signal recovery from random measurements via orthogonal matching pursuit IEEE Trans. Inform. Theory 2007 53 4655 66

[32] Brunton S. L.,Proctor J. L.,Kutz J. N.. Discovering governing equations from data by sparse identification of nonlinear dynamical systems Proc. Natl. Acad. Sci. USA 2016 113 3932 7 PMC4839439

[33] Efron B.,Hastie T.,Johnstone I.,Tibshirani R.. Least angle regression Ann. Statist. 2004 32 407 99

[34] Friedman J.,Hastie T.,Tibshirani R.. Regularization paths for generalized linear models via coordinate descent J. Stat. Soft. 2010 33 1 22 PMC2929880

[35] Zou H.. The adaptive lasso and its oracle properties J. Am. Stat. Assoc. 2012 101 1418 29

[36] Bickel P. J.,Ritov Y.,Tsybakov A. B.. Simultaneous analysis of Lasso and Dantzig selector Ann. Statist. 2009 37 1705 32

[37] Rudy S. H.,Brunton S. L.,Proctor J. L.,Kutz J. N.. Data-driven discovery of partial differential equations Sci. Adv. 2017 3 e1602614

[38] Care A.,Carli R.,Libera A. D.,Romeres D.,Pillonetto G.. Kernel methods and gaussian processes for system identification and control: a road map on regularized kernel-based learning for control IEEE Control Syst. Mag. 2023 43 69 110

[39] Widodo A.,Kim E. Y.,Son J..et al. Fault diagnosis of low speed bearing based on relevance vector machine and support vector machine Expert Syst. Appl. 2009 36 7252 61

[40] Zaman B.,Mckee M.. Spatio-temporal prediction of root zone soil moisture using multivariate relevance vector machines Open J. Mod. Hydrol. 2014 4 80 90

[41] Wei Z.,Yuan W.,Li S.,Yuan J.,Ng D. W. K.. Off-grid channel estimation with sparse Bayesian learning for OTFS systems IEEE Trans. Wireless Commun. 2022 21 7407 26

[42] Wang J.,Zhu Q.,Lin Z..et al. Sparse Bayesian learning-based 3-D radio environment map construction-Sampling optimization, scenario-dependent dictionary construction, and sparse recovery IEEE Trans. Cogn. Commun. Netw. 2024 10 80 93

[43] Wang W.,Qi F.,Wipf D. P..et al. Sparse Bayesian learning for end-to-end EEG decoding IEEE Trans. Pattern Anal. Mach. Intell. 2023 45 15632 49

[44] Wang Q.,Dai Y.,Ma Z..et al. Towards high-precision data modeling of SHM measurements using an improved sparse Bayesian learning scheme with strong generalization ability Struct. Health Monit. 2023 23 588 604

[45] Fang J.,Shen Y.,Li H.,Wang P.. Pattern-coupled sparse Bayesian learning for recovery of block-sparse signals IEEE Trans. Signal Proc. 2014 63 360 72

[46] Tian C.,Dong Z.,Ji Y.,Lin X.. Auxiliary model gradient-based iterative identification with moving data window for Wiener nonlinear output-error systems J. Frankl. Inst. 2025 362 107992

[47] Ding F.,Xiao Y.,Xu L.,Fang Z.. Hierarchical stochastic gradient and hierarchical multi-innovation stochastic gradient identification for multivariable ARX models Int. J. Adapt. Control Signal Process. 2025 40 79 102

[48] Ding F.,Luan X.,Xu L.,Zhang X.. Hierarchical recursive gradient parameter identification for multi-input ARX systems with partially-coupled information vectors Int. J. Adapt. Control Signal Process. 2025 39 1978 95

[49] Xu L.,Ding F.,Zhang X.,Zhu Q.. Novel parameter estimation method for the systems with colored noises by using the filtering identification idea Syst. Control Lett. 2024 186 105774

[50] Ding F.,Xu L.,Zhang X.,Xu H.,Zhou Y.,Luan X.. Hierarchical generalized extended parameter identification for multivariable equation-error ARMA-like systems by using the filtering identification idea Ann. Rev. Control 2025 60 100993

[51] Bengio Y.. Learning deep architectures for AI Found. Trends Mach. Learn. 2009 2 1 127

[52] Wu R.,Jahanshahi M. R.. Deep convolutional neural network for structural dynamic response estimation and system identification J. Eng. Mech. 2019 145 04018125

[53] Wang H.,Luo C.,Wang X.. Synchronization and identification of nonlinear systems by using a novel self-evolving interval type-2 fuzzy LSTM-neural network Eng. Appl. Artif. Intell. 2019 81 79 93

[54] Mangan N. M.,Kutz J. N.,Brunton S. L.,Proctor J. L.. Model selection for dynamical systems via sparse regression and information criteria Proc. R. Soc. A. Math. Phys. Eng. Sci. 2017 473 20170009

[55] Schaeffer H.,Mccalla S. G.. Sparse model selection via integral terms Phys. Rev. E. 2017 96 023302

[56] Fasel U.,Kutz J. N.,Brunton B. W.,Brunton S. L.. Ensemble-SINDy: robust sparse model discovery in the low-data, high-noise limit, with active learning and control Proc. R. Soc. A. 2022 478 20210904 PMC9006119

[57] Messenger D. A.,Bortz D. M.. Weak SINDy for partial differential equations J. Comput. Phys. 2021 443 110525

[58] Xu H.,Chang H.,Zhang D.. DLGA-PDE: discovery of PDEs with incomplete candidate library via combination of deep learning and genetic algorithm J. Comput. Phys. 2020 418 109584

[59] Chen Y.,Luo Y.,Liu Q.,Xu H.,Zhang D.. Symbolic genetic algorithm for discovering open-form partial differential equations (SGA-PDE) Phys. Rev. Res. 2022 4 023174

[60] Zhang X.,Guan J.,Liu Y.,Wang G.. MORL4PDEs: data-driven discovery of PDEs based on multi-objective optimization and reinforcement learning Chaos Soliton Fract. 2024 180 114536

[61] Both G.,Choudhury S.,Sens P.,Kusters R.. DeepMoD: deep learning for model discovery in noisy data J. Comput. Phys. 2021 428 109985

[62] Chen Z.,Liu Y.,Sun H.. Physics-informed learning of governing equations from scarce data Nat. Commun. 2021 12 6136 PMC8531004

[63] Stephany R.,Earls C.. PDE-LEARN: using deep learning to discover partial differential equations from noisy, limited data Neural Netw. 2024 174 106242

[64] Timme M.,Casadiego J.. Revealing networks from dynamics: an introduction J. Phys. A Math. Theor. 2014 47 343001

[65] Han X.,Shen Z.,Wang W.,Lai Y.,Grebogi C.. Reconstructing direct and indirect interactions in networked public goods game Sci. Rep. 2016 6 30241 PMC4996070

[66] Yuan Y.,Tang X.,Zhou W..et al. Data driven discovery of cyber physical systems Nat. Commun. 2019 10 4894 PMC6814766

[67] Kaiser E.,Kutz J. N.,Brunton S. L.. Sparse identification of nonlinear dynamics for model predictive control in the low-data limit Proc. R. Soc. A. 2018 474 20180305

[68] Deng N.,Noack B. R.,Morzyński M.,Pastur L. R.. Galerkin force model for transient and post-transient dynamics of the fluidic pinball J. Fluid Mech. 2021 918 A4

[69] Kaptanoglu A. A.,Morgan K. D.,Hansen C. J.,Brunton S. L.. Physics-constrained, low-dimensional models for magnetohydrodynamics: first-principles and data-driven approaches Phys. Rev. E 2021 104 015206

[70] Sorokina M.,Sygletos S.,Turitsyn S.. Sparse identification for nonlinear optical communication systems: SINO method Opt. Express 2016 24 30433 43

[71] Boninsegna L.,Nüske F.,Clementi C.. Sparse learning of stochastic dynamical equations J. Chem. Phys. 2018 148 241723

[72] Grayeli A.,Sehgal A.,Costilla-Reyes O.,Cranmer M.,Chaudhuri S.. Symbolic regression with a learned concept library Adv. Neural Inf. Proc. Syst. 2024 37 44678 709

[73] Zhang Y.,Zheng K.,Liu F.,Zhang Q.,Wang Z.. AutoTurb: using large language models for automatic algebraic turbulence model discovery Phys. Fluids 2025 37 015211

[74] Holt S.,Qian Z.,Liu T.,Weatherall J.,van der Schaar M.. Data-driven discovery of dynamical systems in pharmacology using large language models Adv. Neural Inf. Process. Syst. 2024 37 96325 66

[75] Spillias S.,Rogers J.,Boschetti F..et al. Data-driven discovery of mechanistic ecosystem models with LLMs bioRxiv 2025 2025 2025 7

[76] Kramer S.,Cerrato M.,Brugger J.,Džeroski S.,King R. D.. Automated scientific discovery: from equation discovery to autonomous discovery systems Mach. Learn. 2026 115 109

[77] Taskin B.,Xie W.,Lazebnik T.. Knowledge integration for physics-informed symbolic regression using pre-trained large language models Sci. Rep. 2026 16 1614 PMC12800073