Matches in SemOpenAlex for { <https://semopenalex.org/work/W4328107281> ?p ?o ?g. }
- W4328107281 endingPage "402" @default.
- W4328107281 startingPage "402" @default.
- W4328107281 abstract "Cascaded control structures are prevalent in industrial systems with many disturbances to obtain stable control but are cumbersome and challenging to tune. In this work, we propose cascaded constrained residual reinforcement learning (RL), an intuitive method that allows to improve the performance of a cascaded control structure while maintaining safe operation at all times. We draw inspiration from the constrained residual RL framework, in which a constrained reinforcement learning agent learns corrective adaptations to a base controller’s output to increase optimality. We first revisit the interplay between the residual agent and the baseline controller and subsequently extend this to the cascaded case. We analyze the differences and challenges this structure brings and derive some principle insights from this into the stability and operation of the cascaded residual architecture. Next, we propose a novel actor structure to enable efficient learning under the cascaded setting. We show that the standard algorithm is suboptimal for application to cascaded control structures and validate our method on a high-fidelity simulator of a dual motor drivetrain, resulting in a performance improvement of 14.7% on average, with only a minor decrease in performance occurring during the training phase. We study the different principles constituting the method and examine and validate their contribution to the algorithm’s performance under the considered cascaded control structure." @default.
- W4328107281 created "2023-03-22" @default.
- W4328107281 creator A5008049481 @default.
- W4328107281 creator A5058733585 @default.
- W4328107281 creator A5067813287 @default.
- W4328107281 date "2023-03-20" @default.
- W4328107281 modified "2023-10-18" @default.
- W4328107281 title "Optimizing Cascaded Control of Mechatronic Systems through Constrained Residual Reinforcement Learning" @default.
- W4328107281 cites W2035807953 @default.
- W4328107281 cites W2038051031 @default.
- W4328107281 cites W2049655011 @default.
- W4328107281 cites W2065927905 @default.
- W4328107281 cites W2347203552 @default.
- W4328107281 cites W2745496942 @default.
- W4328107281 cites W2766447205 @default.
- W4328107281 cites W2794566841 @default.
- W4328107281 cites W2948652605 @default.
- W4328107281 cites W2985713881 @default.
- W4328107281 cites W2996488578 @default.
- W4328107281 cites W3021849639 @default.
- W4328107281 cites W3037533914 @default.
- W4328107281 cites W3044110240 @default.
- W4328107281 cites W3100789280 @default.
- W4328107281 cites W3168892396 @default.
- W4328107281 cites W3183327882 @default.
- W4328107281 cites W3202294003 @default.
- W4328107281 cites W3202600213 @default.
- W4328107281 cites W4206562247 @default.
- W4328107281 cites W4213377513 @default.
- W4328107281 cites W4214668146 @default.
- W4328107281 cites W4224103873 @default.
- W4328107281 cites W4224268942 @default.
- W4328107281 cites W4292070354 @default.
- W4328107281 cites W4293152416 @default.
- W4328107281 cites W4310705874 @default.
- W4328107281 cites W4322100937 @default.
- W4328107281 doi "https://doi.org/10.3390/machines11030402" @default.
- W4328107281 hasPublicationYear "2023" @default.
- W4328107281 type Work @default.
- W4328107281 citedByCount "0" @default.
- W4328107281 crossrefType "journal-article" @default.
- W4328107281 hasAuthorship W4328107281A5008049481 @default.
- W4328107281 hasAuthorship W4328107281A5058733585 @default.
- W4328107281 hasAuthorship W4328107281A5067813287 @default.
- W4328107281 hasBestOaLocation W43281072811 @default.
- W4328107281 hasConcept C112972136 @default.
- W4328107281 hasConcept C11413529 @default.
- W4328107281 hasConcept C119857082 @default.
- W4328107281 hasConcept C127413603 @default.
- W4328107281 hasConcept C133731056 @default.
- W4328107281 hasConcept C154945302 @default.
- W4328107281 hasConcept C155512373 @default.
- W4328107281 hasConcept C203479927 @default.
- W4328107281 hasConcept C2775924081 @default.
- W4328107281 hasConcept C2776459999 @default.
- W4328107281 hasConcept C28704281 @default.
- W4328107281 hasConcept C41008148 @default.
- W4328107281 hasConcept C47446073 @default.
- W4328107281 hasConcept C6557445 @default.
- W4328107281 hasConcept C76155785 @default.
- W4328107281 hasConcept C86803240 @default.
- W4328107281 hasConcept C97541855 @default.
- W4328107281 hasConceptScore W4328107281C112972136 @default.
- W4328107281 hasConceptScore W4328107281C11413529 @default.
- W4328107281 hasConceptScore W4328107281C119857082 @default.
- W4328107281 hasConceptScore W4328107281C127413603 @default.
- W4328107281 hasConceptScore W4328107281C133731056 @default.
- W4328107281 hasConceptScore W4328107281C154945302 @default.
- W4328107281 hasConceptScore W4328107281C155512373 @default.
- W4328107281 hasConceptScore W4328107281C203479927 @default.
- W4328107281 hasConceptScore W4328107281C2775924081 @default.
- W4328107281 hasConceptScore W4328107281C2776459999 @default.
- W4328107281 hasConceptScore W4328107281C28704281 @default.
- W4328107281 hasConceptScore W4328107281C41008148 @default.
- W4328107281 hasConceptScore W4328107281C47446073 @default.
- W4328107281 hasConceptScore W4328107281C6557445 @default.
- W4328107281 hasConceptScore W4328107281C76155785 @default.
- W4328107281 hasConceptScore W4328107281C86803240 @default.
- W4328107281 hasConceptScore W4328107281C97541855 @default.
- W4328107281 hasIssue "3" @default.
- W4328107281 hasLocation W43281072811 @default.
- W4328107281 hasLocation W43281072812 @default.
- W4328107281 hasLocation W43281072813 @default.
- W4328107281 hasLocation W43281072814 @default.
- W4328107281 hasOpenAccess W4328107281 @default.
- W4328107281 hasPrimaryLocation W43281072811 @default.
- W4328107281 hasRelatedWork W2029495890 @default.
- W4328107281 hasRelatedWork W2047787160 @default.
- W4328107281 hasRelatedWork W2099111500 @default.
- W4328107281 hasRelatedWork W2152858547 @default.
- W4328107281 hasRelatedWork W2365785480 @default.
- W4328107281 hasRelatedWork W2393509435 @default.
- W4328107281 hasRelatedWork W2909158123 @default.
- W4328107281 hasRelatedWork W2959276766 @default.
- W4328107281 hasRelatedWork W3074294383 @default.
- W4328107281 hasRelatedWork W4206669594 @default.
- W4328107281 hasVolume "11" @default.
- W4328107281 isParatext "false" @default.