Unifying Model-Based and Neural Network Feedforward: Physics-Guided Neural Networks with Linear Autoregressive Dynamics
Unknown nonlinear dynamics often limit the tracking performance of feedforward control. The aim of this paper is to develop a feedforward control framework that can compensate these unknown nonlinear dynamics using universal function approximators. The feedforward controller is parametrized as a par...
Gespeichert in:
| Veröffentlicht in: | arXiv.org 2022-09 |
|---|---|
| Hauptverfasser: | , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | arXiv.org |
| container_volume | |
| creator | Kon, Johan Bruijnen, Dennis van de Wijdeven, Jeroen Heertjes, Marcel Oomen, Tom |
| description | Unknown nonlinear dynamics often limit the tracking performance of feedforward control. The aim of this paper is to develop a feedforward control framework that can compensate these unknown nonlinear dynamics using universal function approximators. The feedforward controller is parametrized as a parallel combination of a physics-based model and a neural network, where both share the same linear autoregressive (AR) dynamics. This parametrization allows for efficient output-error optimization through Sanathanan-Koerner (SK) iterations. Within each SK-iteration, the output of the neural network is penalized in the subspace of the physics-based model through orthogonal projection-based regularization, such that the neural network captures only the unmodelled dynamics, resulting in interpretable models. |
| doi_str_mv | 10.48550/arxiv.2209.12489 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_2209_12489</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2718478068</sourcerecordid><originalsourceid>FETCH-LOGICAL-a959-db37024fa2f86201865559ac25f5e47220fe37cf22af35598716354f589f75d23</originalsourceid><addsrcrecordid>eNpdkE1TwjAQhjPOOCOD_ABPZsZzMd00TeoNUdAZ_DjguRPJBoLQYtJS-u-t4MnTe9jn3dl9CLmK2TBRQrBb7Q9uPwRg2TCGRGVnpAecx5FKAC7IIIQ1YwxSCULwHmk-CmdbVyzpS2lwE93rgIbqwtBXrL3edFE1pf-iE0RjS99ob-7o-6oNbhGiae0M_kcDbVy1ojNXoPZ0VFelx6XHENwe6UNb6G1XvSTnVm8CDv6yT-aTx_n4KZq9TZ_Ho1mkM5FF5pNLBonVYFUKLFapECLTCxBWYCK7Jy1yubAA2vJuomSccpFYoTIrhQHeJ9entUcr-c67rfZt_msnP9rpiJsTsfPld42hytdl7YvuphxkrBKpWKr4D1MCaC8</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2718478068</pqid></control><display><type>article</type><title>Unifying Model-Based and Neural Network Feedforward: Physics-Guided Neural Networks with Linear Autoregressive Dynamics</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Kon, Johan ; Bruijnen, Dennis ; van de Wijdeven, Jeroen ; Heertjes, Marcel ; Oomen, Tom</creator><creatorcontrib>Kon, Johan ; Bruijnen, Dennis ; van de Wijdeven, Jeroen ; Heertjes, Marcel ; Oomen, Tom</creatorcontrib><description>Unknown nonlinear dynamics often limit the tracking performance of feedforward control. The aim of this paper is to develop a feedforward control framework that can compensate these unknown nonlinear dynamics using universal function approximators. The feedforward controller is parametrized as a parallel combination of a physics-based model and a neural network, where both share the same linear autoregressive (AR) dynamics. This parametrization allows for efficient output-error optimization through Sanathanan-Koerner (SK) iterations. Within each SK-iteration, the output of the neural network is penalized in the subspace of the physics-based model through orthogonal projection-based regularization, such that the neural network captures only the unmodelled dynamics, resulting in interpretable models.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.2209.12489</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Autoregressive processes ; Computer Science - Learning ; Computer Science - Systems and Control ; Feedforward control ; Iterative methods ; Neural networks ; Nonlinear dynamics ; Optimization ; Parameterization ; Regularization ; Tracking control</subject><ispartof>arXiv.org, 2022-09</ispartof><rights>2022. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,777,781,882,27906</link.rule.ids><backlink>$$Uhttps://doi.org/10.48550/arXiv.2209.12489$$DView paper in arXiv$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.1109/CDC51059.2022.9992511$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink></links><search><creatorcontrib>Kon, Johan</creatorcontrib><creatorcontrib>Bruijnen, Dennis</creatorcontrib><creatorcontrib>van de Wijdeven, Jeroen</creatorcontrib><creatorcontrib>Heertjes, Marcel</creatorcontrib><creatorcontrib>Oomen, Tom</creatorcontrib><title>Unifying Model-Based and Neural Network Feedforward: Physics-Guided Neural Networks with Linear Autoregressive Dynamics</title><title>arXiv.org</title><description>Unknown nonlinear dynamics often limit the tracking performance of feedforward control. The aim of this paper is to develop a feedforward control framework that can compensate these unknown nonlinear dynamics using universal function approximators. The feedforward controller is parametrized as a parallel combination of a physics-based model and a neural network, where both share the same linear autoregressive (AR) dynamics. This parametrization allows for efficient output-error optimization through Sanathanan-Koerner (SK) iterations. Within each SK-iteration, the output of the neural network is penalized in the subspace of the physics-based model through orthogonal projection-based regularization, such that the neural network captures only the unmodelled dynamics, resulting in interpretable models.</description><subject>Autoregressive processes</subject><subject>Computer Science - Learning</subject><subject>Computer Science - Systems and Control</subject><subject>Feedforward control</subject><subject>Iterative methods</subject><subject>Neural networks</subject><subject>Nonlinear dynamics</subject><subject>Optimization</subject><subject>Parameterization</subject><subject>Regularization</subject><subject>Tracking control</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GOX</sourceid><recordid>eNpdkE1TwjAQhjPOOCOD_ABPZsZzMd00TeoNUdAZ_DjguRPJBoLQYtJS-u-t4MnTe9jn3dl9CLmK2TBRQrBb7Q9uPwRg2TCGRGVnpAecx5FKAC7IIIQ1YwxSCULwHmk-CmdbVyzpS2lwE93rgIbqwtBXrL3edFE1pf-iE0RjS99ob-7o-6oNbhGiae0M_kcDbVy1ojNXoPZ0VFelx6XHENwe6UNb6G1XvSTnVm8CDv6yT-aTx_n4KZq9TZ_Ho1mkM5FF5pNLBonVYFUKLFapECLTCxBWYCK7Jy1yubAA2vJuomSccpFYoTIrhQHeJ9entUcr-c67rfZt_msnP9rpiJsTsfPld42hytdl7YvuphxkrBKpWKr4D1MCaC8</recordid><startdate>20220926</startdate><enddate>20220926</enddate><creator>Kon, Johan</creator><creator>Bruijnen, Dennis</creator><creator>van de Wijdeven, Jeroen</creator><creator>Heertjes, Marcel</creator><creator>Oomen, Tom</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20220926</creationdate><title>Unifying Model-Based and Neural Network Feedforward: Physics-Guided Neural Networks with Linear Autoregressive Dynamics</title><author>Kon, Johan ; Bruijnen, Dennis ; van de Wijdeven, Jeroen ; Heertjes, Marcel ; Oomen, Tom</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a959-db37024fa2f86201865559ac25f5e47220fe37cf22af35598716354f589f75d23</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Autoregressive processes</topic><topic>Computer Science - Learning</topic><topic>Computer Science - Systems and Control</topic><topic>Feedforward control</topic><topic>Iterative methods</topic><topic>Neural networks</topic><topic>Nonlinear dynamics</topic><topic>Optimization</topic><topic>Parameterization</topic><topic>Regularization</topic><topic>Tracking control</topic><toplevel>online_resources</toplevel><creatorcontrib>Kon, Johan</creatorcontrib><creatorcontrib>Bruijnen, Dennis</creatorcontrib><creatorcontrib>van de Wijdeven, Jeroen</creatorcontrib><creatorcontrib>Heertjes, Marcel</creatorcontrib><creatorcontrib>Oomen, Tom</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection><collection>arXiv Computer Science</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kon, Johan</au><au>Bruijnen, Dennis</au><au>van de Wijdeven, Jeroen</au><au>Heertjes, Marcel</au><au>Oomen, Tom</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Unifying Model-Based and Neural Network Feedforward: Physics-Guided Neural Networks with Linear Autoregressive Dynamics</atitle><jtitle>arXiv.org</jtitle><date>2022-09-26</date><risdate>2022</risdate><eissn>2331-8422</eissn><abstract>Unknown nonlinear dynamics often limit the tracking performance of feedforward control. The aim of this paper is to develop a feedforward control framework that can compensate these unknown nonlinear dynamics using universal function approximators. The feedforward controller is parametrized as a parallel combination of a physics-based model and a neural network, where both share the same linear autoregressive (AR) dynamics. This parametrization allows for efficient output-error optimization through Sanathanan-Koerner (SK) iterations. Within each SK-iteration, the output of the neural network is penalized in the subspace of the physics-based model through orthogonal projection-based regularization, such that the neural network captures only the unmodelled dynamics, resulting in interpretable models.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.2209.12489</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | EISSN: 2331-8422 |
| ispartof | arXiv.org, 2022-09 |
| issn | 2331-8422 |
| language | eng |
| recordid | cdi_arxiv_primary_2209_12489 |
| source | arXiv.org; Free E- Journals |
| subjects | Autoregressive processes Computer Science - Learning Computer Science - Systems and Control Feedforward control Iterative methods Neural networks Nonlinear dynamics Optimization Parameterization Regularization Tracking control |
| title | Unifying Model-Based and Neural Network Feedforward: Physics-Guided Neural Networks with Linear Autoregressive Dynamics |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-21T06%3A05%3A18IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Unifying%20Model-Based%20and%20Neural%20Network%20Feedforward:%20Physics-Guided%20Neural%20Networks%20with%20Linear%20Autoregressive%20Dynamics&rft.jtitle=arXiv.org&rft.au=Kon,%20Johan&rft.date=2022-09-26&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.2209.12489&rft_dat=%3Cproquest_arxiv%3E2718478068%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2718478068&rft_id=info:pmid/&rfr_iscdi=true |