Robust Model Predictive Control of Nonlinear Systems With Unmodeled Dynamics and Bounded Uncertainties Based on Neural Networks. Yan, Z. & Wang, J. IEEE Transactions on Neural Networks and Learning Systems, 25(3):457–469, March, 2014. Conference Name: IEEE Transactions on Neural Networks and Learning Systemsdoi abstract bibtex This paper presents a neural network approach to robust model predictive control (MPC) for constrained discrete-time nonlinear systems with unmodeled dynamics affected by bounded uncertainties. The exact nonlinear model of underlying process is not precisely known, but a partially known nominal model is available. This partially known nonlinear model is first decomposed to an affine term plus an unknown high-order term via Jacobian linearization. The linearization residue combined with unmodeled dynamics is then modeled using an extreme learning machine via supervised learning. The minimax methodology is exploited to deal with bounded uncertainties. The minimax optimization problem is reformulated as a convex minimization problem and is iteratively solved by a two-layer recurrent neural network. The proposed neurodynamic approach to nonlinear MPC improves the computational efficiency and sheds a light for real-time implementability of MPC technology. Simulation results are provided to substantiate the effectiveness and characteristics of the proposed approach.
@article{yan_robust_2014,
title = {Robust {Model} {Predictive} {Control} of {Nonlinear} {Systems} {With} {Unmodeled} {Dynamics} and {Bounded} {Uncertainties} {Based} on {Neural} {Networks}},
volume = {25},
issn = {2162-2388},
doi = {10.1109/tnnls.2013.2275948},
abstract = {This paper presents a neural network approach to robust model predictive control (MPC) for constrained discrete-time nonlinear systems with unmodeled dynamics affected by bounded uncertainties. The exact nonlinear model of underlying process is not precisely known, but a partially known nominal model is available. This partially known nonlinear model is first decomposed to an affine term plus an unknown high-order term via Jacobian linearization. The linearization residue combined with unmodeled dynamics is then modeled using an extreme learning machine via supervised learning. The minimax methodology is exploited to deal with bounded uncertainties. The minimax optimization problem is reformulated as a convex minimization problem and is iteratively solved by a two-layer recurrent neural network. The proposed neurodynamic approach to nonlinear MPC improves the computational efficiency and sheds a light for real-time implementability of MPC technology. Simulation results are provided to substantiate the effectiveness and characteristics of the proposed approach.},
number = {3},
journal = {IEEE Transactions on Neural Networks and Learning Systems},
author = {Yan, Zheng and Wang, Jun},
month = mar,
year = {2014},
note = {Conference Name: IEEE Transactions on Neural Networks and Learning Systems},
keywords = {/unread, Artificial neural networks, Extreme learning machine (ELM), Nonlinear systems, Optimization, Robustness, Uncertainty, Vectors, real-time optimization, recurrent neural networks (RNNs), robust model predictive control (MPC), unmodeled dynamics},
pages = {457--469},
}
Downloads: 0
{"_id":"p99Nk4YpptGv3DgHw","bibbaseid":"yan-wang-robustmodelpredictivecontrolofnonlinearsystemswithunmodeleddynamicsandboundeduncertaintiesbasedonneuralnetworks-2014","author_short":["Yan, Z.","Wang, J."],"bibdata":{"bibtype":"article","type":"article","title":"Robust Model Predictive Control of Nonlinear Systems With Unmodeled Dynamics and Bounded Uncertainties Based on Neural Networks","volume":"25","issn":"2162-2388","doi":"10.1109/tnnls.2013.2275948","abstract":"This paper presents a neural network approach to robust model predictive control (MPC) for constrained discrete-time nonlinear systems with unmodeled dynamics affected by bounded uncertainties. The exact nonlinear model of underlying process is not precisely known, but a partially known nominal model is available. This partially known nonlinear model is first decomposed to an affine term plus an unknown high-order term via Jacobian linearization. The linearization residue combined with unmodeled dynamics is then modeled using an extreme learning machine via supervised learning. The minimax methodology is exploited to deal with bounded uncertainties. The minimax optimization problem is reformulated as a convex minimization problem and is iteratively solved by a two-layer recurrent neural network. The proposed neurodynamic approach to nonlinear MPC improves the computational efficiency and sheds a light for real-time implementability of MPC technology. Simulation results are provided to substantiate the effectiveness and characteristics of the proposed approach.","number":"3","journal":"IEEE Transactions on Neural Networks and Learning Systems","author":[{"propositions":[],"lastnames":["Yan"],"firstnames":["Zheng"],"suffixes":[]},{"propositions":[],"lastnames":["Wang"],"firstnames":["Jun"],"suffixes":[]}],"month":"March","year":"2014","note":"Conference Name: IEEE Transactions on Neural Networks and Learning Systems","keywords":"/unread, Artificial neural networks, Extreme learning machine (ELM), Nonlinear systems, Optimization, Robustness, Uncertainty, Vectors, real-time optimization, recurrent neural networks (RNNs), robust model predictive control (MPC), unmodeled dynamics","pages":"457–469","bibtex":"@article{yan_robust_2014,\n\ttitle = {Robust {Model} {Predictive} {Control} of {Nonlinear} {Systems} {With} {Unmodeled} {Dynamics} and {Bounded} {Uncertainties} {Based} on {Neural} {Networks}},\n\tvolume = {25},\n\tissn = {2162-2388},\n\tdoi = {10.1109/tnnls.2013.2275948},\n\tabstract = {This paper presents a neural network approach to robust model predictive control (MPC) for constrained discrete-time nonlinear systems with unmodeled dynamics affected by bounded uncertainties. The exact nonlinear model of underlying process is not precisely known, but a partially known nominal model is available. This partially known nonlinear model is first decomposed to an affine term plus an unknown high-order term via Jacobian linearization. The linearization residue combined with unmodeled dynamics is then modeled using an extreme learning machine via supervised learning. The minimax methodology is exploited to deal with bounded uncertainties. The minimax optimization problem is reformulated as a convex minimization problem and is iteratively solved by a two-layer recurrent neural network. The proposed neurodynamic approach to nonlinear MPC improves the computational efficiency and sheds a light for real-time implementability of MPC technology. Simulation results are provided to substantiate the effectiveness and characteristics of the proposed approach.},\n\tnumber = {3},\n\tjournal = {IEEE Transactions on Neural Networks and Learning Systems},\n\tauthor = {Yan, Zheng and Wang, Jun},\n\tmonth = mar,\n\tyear = {2014},\n\tnote = {Conference Name: IEEE Transactions on Neural Networks and Learning Systems},\n\tkeywords = {/unread, Artificial neural networks, Extreme learning machine (ELM), Nonlinear systems, Optimization, Robustness, Uncertainty, Vectors, real-time optimization, recurrent neural networks (RNNs), robust model predictive control (MPC), unmodeled dynamics},\n\tpages = {457--469},\n}\n\n","author_short":["Yan, Z.","Wang, J."],"key":"yan_robust_2014","id":"yan_robust_2014","bibbaseid":"yan-wang-robustmodelpredictivecontrolofnonlinearsystemswithunmodeleddynamicsandboundeduncertaintiesbasedonneuralnetworks-2014","role":"author","urls":{},"keyword":["/unread","Artificial neural networks","Extreme learning machine (ELM)","Nonlinear systems","Optimization","Robustness","Uncertainty","Vectors","real-time optimization","recurrent neural networks (RNNs)","robust model predictive control (MPC)","unmodeled dynamics"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/victorjhu","dataSources":["CmHEoydhafhbkXXt5"],"keywords":["/unread","artificial neural networks","extreme learning machine (elm)","nonlinear systems","optimization","robustness","uncertainty","vectors","real-time optimization","recurrent neural networks (rnns)","robust model predictive control (mpc)","unmodeled dynamics"],"search_terms":["robust","model","predictive","control","nonlinear","systems","unmodeled","dynamics","bounded","uncertainties","based","neural","networks","yan","wang"],"title":"Robust Model Predictive Control of Nonlinear Systems With Unmodeled Dynamics and Bounded Uncertainties Based on Neural Networks","year":2014}