Identification of systems with unknown inputs using indirect input measurements

ABSTRACT A common issue with many system identification problems is that the true input to the system is unknown. This paper extends a previously presented indirect modelling framework that deals with identification of systems where the input is partially or fully unknown. In this framework, unknown inputs are eliminated by using additional measurements that directly or indirectly contain information about the unknown inputs. The resulting indirect predictor model is only dependent on known and measured signals and can be used to estimate the desired dynamics or properties. Since the input of the indirect model contains both known inputs and measurements that could all be correlated with the same disturbances as the output, estimation of the indirect model has similar challenges as a closed-loop estimation problem. In fact, due to the generality of the indirect modelling framework, it unifies a number of already existing system identification problems that are contained as special cases. For completeness, the paper is concluded with one method that can be used to estimate the indirect model as well as an experimental verification to show the applicability of the framework.

[1]  Hugues Garnier,et al.  Identification of continuous-time errors-in-variables models , 2006, Autom..

[2]  Martin Enqvist,et al.  Estimating models of inverse systems , 2013, 52nd IEEE Conference on Decision and Control.

[3]  Eric Walter,et al.  Identifiability of State Space Models , 1982 .

[4]  T. M. Cannon,et al.  Blind deconvolution through digital signal processing , 1975, Proceedings of the IEEE.

[5]  Dennis S. Bernstein,et al.  Sensor-Only Noncausal Blind Identification of Pseudo Transfer Functions , 2009 .

[6]  P. Müller,et al.  Disturbance decoupled observer design: a unified viewpoint , 1994, IEEE Trans. Autom. Control..

[7]  Xavier Bombois,et al.  Identification of dynamic models in complex networks with prediction error methods - Basic methods for consistent module estimates , 2013, Autom..

[8]  Peter C. Young,et al.  The advantages of directly identifying continuous-time transfer function models in practical applications , 2014, Int. J. Control.

[9]  Rolf Johansson.  System Identification by T. Söderström and P. Stoica , 1994 .

[10]  Lennart Ljung,et al.  System identification (2nd ed.): theory for the user , 1999 .

[11]  Fredrik Gustafsson,et al.  Online Estimation of Ship's Mass and Center of Mass Using Inertial Measurements , 2015 .

[12]  Alexandre S. Bazanella,et al.  Closed-loop identification of MIMO systems: a new look at identifiability and experiment design , 2007, 2007 European Control Conference (ECC).

[13]  Alex Simpkins,et al.  System Identification: Theory for the User, 2nd Edition (Ljung, L.; 1999) [On the Shelf] , 2012, IEEE Robotics & Automation Magazine.

[14]  Arne G. Dankers.  System Identification in Dynamic Networks , 2014 .

[15]  U. Forssell,et al.  Efficiency of prediction error and instrumental variable methods for closed-loop identification , 1998, Proceedings of the 37th IEEE Conference on Decision and Control (Cat. No.98CH36171).

[16]  M. Anguelova.  Observability and identifiability of nonlinear systems with applications in biology , 2007 .

[17]  Torsten Söderström,et al.  Errors-in-variables methods in system identification , 2018, Autom..

[18]  Fredrik Gustafsson,et al.  Modeling for IMU-based Online Estimation of a Ship's Mass and Center of Mass , 2015 .

[19]  Robert H. Shumway,et al.  Time series analysis and its applications : with R examples , 2017 .

[20]  M. Fliess,et al.  Nonlinear observability, identifiability, and persistent trajectories , 1991, [1991] Proceedings of the 30th IEEE Conference on Decision and Control.

[21]  R. Patton,et al.  Optimal filtering for systems with unknown inputs , 1998, IEEE Trans. Autom. Control..

[22]  Arun K. Tangirala,et al.  Principles of System Identification , 2014 .

[23]  Petre Stoica,et al.  Decentralized Control , 2018, The Control Systems Handbook.

[24]  Arild Thowsen,et al.  Structural identifiability , 1977, 1977 IEEE Conference on Decision and Control including the 16th Symposium on Adaptive Processes and A Special Symposium on Fuzzy Set Theory and Applications.

[25]  Torsten Söderström,et al.  Can errors-in-variables systems be identified from closed-loop experiments? , 2013, Autom..

[26]  Liuping Wang,et al.  Identification of Continuous-time Models from Sampled Data , 2008 .

[27]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[28]  Lennart Ljung,et al.  Closed-loop identification revisited , 1999, Autom..

[29]  A. Krener,et al.  Nonlinear controllability and observability , 1977 .

[30]  Lennart Ljung,et al.  On global identifiability for arbitrary model parametrizations , 1994, Autom..

[31]  Karim Abed-Meraim,et al.  Blind system identification , 1997, Proc. IEEE.

[32]  Marion Gilson,et al.  Instrumental variable methods for closed-loop system identification , 2005, Autom..

[33]  Dennis S. Bernstein,et al.  Sensor-to-sensor identification of Hammerstein systems , 2012, 2012 IEEE 51st IEEE Conference on Decision and Control (CDC).

[34]  Fredrik Gustafsson,et al.  A closed-loop instrumental variable approach to mass and center of mass estimation using IMU data , 2014, 53rd IEEE Conference on Decision and Control.

[35]  Fredrik Gustafsson,et al.  Identifiability of physical parameters in systems with limited sensors , 2014 .

[36]  Eric Walter,et al.  Identifiability of State Space Models: with applications to transformation systems , 1982 .

[37]  P. Lancaster,et al.  The theory of matrices : with applications , 1985 .

[38]  P. Young,et al.  Optimal instrumental variable method for closed-loop identification , 2011 .

[39]  Jonas Linder,et al.  On Indirect Input Measurements , 2015 .