In the framework of the subspace-based identification of linear systems, the first step for the construction of a state-space model from observed input-output data involves the estimation of the output predictor. Such construction is based on projection operations of certain structured data matrices onto suitable subspaces spanned by the collected data. To the purpose of predictive control using short-term predictors, this algorithmic step can be elaborated to provide data-based multi-step predictors. Using such an approach, this contribution deals with subspace-based identification methods for the estimation of short-term predictors. One illustrative example is provided: blood glucose prediction in type 1 diabetes mellitus.

Chapter Contents:

• Abstract
• 6.1 Introduction
• 6.1.1 Model description
• 6.1.2 Notation
• 6.1.3 Statement of the problem
• 6.2 Subspace-based linear multi-step predictors
• 6.2.1 Computing projections
• 6.3 Example
• 6.3.1 Diabetes mellitus
• 6.3.2 Experimental conditions
• 6.3.3 Prediction strategy
• 6.3.4 Results
• 6.4 Discussion and conclusions
• References

Inspec keywords: predictive control; linear systems; matrix algebra; identification

Other keywords: short-term predictors; projection operations; type 1 diabetes mellitus; state-space model; short-term predictor estimation; subspace-based multistep predictors; subspace-based identification; blood glucose prediction; linear systems; output predictor estimation; data-based multistep predictors; predictive control; structured data matrices

Subjects: Optimal control; Algebra; Simulation, modelling and identification; Linear control systems