Comparison of Adaptive and Model-Free Methods for Dynamic Measurement
Dynamic measurement aims to improve the speed and accuracy characteristics of measurement devices by signal processing. State-of-the-art dynamic measurement methods are model-based adaptive methods, i.e., 1) they estimate model parameters in real-time and 2) based on the identified model perform model-based signal processing.
The proposed model-free method belongs to the class of the subspace identification methods. It computes directly the quantity of interest without an explicit parameter estimation. This allows efficient computation as well as applicability to general high order multivariable processes.