On system identification of complex systems from finite data

We introduce a new principle for identification based on choosing a model from the model-parameterization, which best approximates the unknown real system belonging to a more complex space of systems that do not lend themselves to a finite-parameterization. The principle is particularly effective for robust control as it leads to a precise notion of parametric and nonparametric error, and the identification problem can be equivalently perceived as that of robust convergence of the parameters in the face of unmodeled errors. The main difficulty in its application stems from the interplay of noise and unmodeled dynamics and requires developing novel two-step algorithms that amount to annihilation of the unmodeled error followed by averaging out the noise. The paper establishes: robust convergence for a large class of systems, topologies, and unmodeled errors; sample path based finite-time polynomial rate of convergence; and annihilation-correlation algorithms for linearly parameterized model structures.

[1]  Pertti M. Mäkilä.  Robust Control-Oriented Identification , 1997 .

[2]  P. M. Makila.  On robust control-oriented identification of discrete and continuous-time systems , 1998 .

[3]  P. Graefe.  Linear stochastic systems , 1966 .

[4]  Kai Lai Chung,et al.  A Course in Probability Theory , 1949 .

[5]  Munther A. Dahleh,et al.  Classical system identification in a deterministic setting , 1995, Proceedings of 1995 34th IEEE Conference on Decision and Control.

[6]  M. Milanese,et al.  Model quality evaluation in H2 identification , 1997, IEEE Trans. Autom. Control..

[7]  H. Woxniakowski.  Information-Based Complexity , 1988 .

[8]  Munther A. Dahleh,et al.  On robust control synthesis and analysis in a Hilbert space , 2000 .

[9]  Munther A. Dahleh,et al.  Classical identification in a deterministic setting , 1995 .

[10]  A. Pinkus.  n-Widths in Approximation Theory , 1985 .

[11]  Munther A. Dahleh,et al.  System identification of complex systems: problem formulation and results , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.

[12]  John N. Tsitsiklis,et al.  Optimal asymptotic identification under bounded disturbances , 1993, IEEE Trans. Autom. Control..

[13]  E. Yaz.  Linear Matrix Inequalities In System And Control Theory , 1998, Proceedings of the IEEE.

[14]  G. Zames.  On the metric complexity of casual linear systems: ε -Entropy and ε -Dimension for continuous time , 1979 .

[15]  Y. F. Huang,et al.  On the value of information in system identification - Bounded noise case , 1982, Autom..

[16]  J. Littlewood,et al.  Some Problems of Diophantine Approximation: The Lattice-Points of a Right-Angled Triangle , 1922 .

[17]  H. T. Shu,et al.  Longitudinal vibration of a hoist rope coupled with the vertical vibration of an elevator car , 1991 .

[18]  Jonathan R. Partington,et al.  On robustness in system identification , 1999, Autom..

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

[20]  Jonathan R. Partington,et al.  Analysis of linear methods for robust identification in ℓ1 , 1995, Autom..

[21]  Saligrama Ramaswamy Venkatesh.  System-identification for complex-systems , 1997 .

[22]  Henryk Wozniakowski,et al.  Information-based complexity , 1987, Nature.

[23]  Michel Gevers,et al.  Towards a Joint Design of Identification and Control , 1993 .

[24]  K. Poolla,et al.  On the time complexity of worst-case system identification , 1994, IEEE Trans. Autom. Control..

[25]  G. Zames.  On the metric complexity of causal linear systems: Estimates of ε-entropy and ε-dimension , 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.

[26]  J. Tsitsiklis,et al.  The sample complexity of worst-case identification of FIR linear systems , 1993, Proceedings of 32nd IEEE Conference on Decision and Control.

[27]  Munther A. Dahleh,et al.  Identification in the presence of classes of unmodeled dynamics and noise , 1997, IEEE Trans. Autom. Control..

[28]  Mathukumalli Vidyasagar,et al.  A Theory of Learning and Generalization: With Applications to Neural Networks and Control Systems , 1997 .

[29]  Y. M. Cho,et al.  Identification and control of high-rise elevators , 1998, Proceedings of the 1998 American Control Conference. ACC (IEEE Cat. No.98CH36207).

[30]  P.M. Mäkilä,et al.  Worst-case control-relevant identification , 1995, Autom..

[31]  L. Ljung.  Convergence analysis of parametric identification methods , 1978 .

[32]  G. H. Hardy,et al.  Some problems of Diophantine approximation: The lattice-points of a right-angled triangle. (Second memoir.) , 1922 .

[33]  Lennart Ljung,et al.  Worst-case identification in l/sub 1/ for f.i.r linear systems , 1995, Proceedings of 1995 34th IEEE Conference on Decision and Control.

[34]  Fernando Paganini,et al.  White Noise Rejection in a Deterministic Setting , 1994 .

[35]  D. Luenberger.  Optimization by Vector Space Methods , 1968 .

[36]  Carl N. Nett,et al.  Control oriented system identification: a worst-case/deterministic approach in H/sub infinity / , 1991 .

[37]  Laura Giarré,et al.  Model quality evaluation in set membership identification , 1997, Autom..

[38]  M. Milanese.  Model Quality Evaluation in Identification , 1997 .

[39]  Antonio Vicino,et al.  Information-Based Complexity and Nonparametric Worst-Case System Identification , 1993, J. Complex..

[40]  E. Engel.  A road to randomness in physical systems , 1992 .