摘要：AbstractThis paper aims at finding a new way to implement subspace-based continuoustime model identification methods in noise corrupted case. A novel nuclear norm method for continuous-time model is first proposed. A connection between the noise constraint and optimization variables is solved during constructing the optimization problem. Finally, a simple numerical simulation is accomplished to show the efficacy of the proposed method.
关键词：KeywordsSystem identificationSubspace methodsContinuous-time modelOptimization problemsSingular value compositionNuclear norm method
