Authors:
Mateus Giesbrecht
;
Gilmar Barreto
and
Celso Pascoli Bottura
Affiliation:
School of Electrical and Computer Engineering FEEC, University of Campinas UNICAMP, Campinas, SP and Brazil
Keyword(s):
Identification Algorithms, Least-squares Identification, Parameter Identification, System Identification, Linear Multivariable Systems, State-space Realization, Impulse Responses, Markov Parameters.
Related
Ontology
Subjects/Areas/Topics:
Informatics in Control, Automation and Robotics
;
Signal Processing, Sensors, Systems Modeling and Control
;
System Identification
;
System Modeling
Abstract:
In this paper the definitions of multivariable discrete impulse and multivariable discrete impulse response are clearly stated and explored. From these definitions, two methods to determine the Markov parameters of a multivariable linear system from input and output data are described. Combining any of the methods to a known method to determine the state space model matrices from the Markov parameters, a practical algorithm to determine state space models from input and output data is obtained. The algorithm is then implemented and compared to a known subspace identification algorithm. The main contributions of this paper are the explicit definitions of multivariable discrete impulse and multivariable discrete impulse response, the discussion of these concepts and the application of them to solve the multivariable linear system identification problem.