Abstract
This chapter introduces the fundamental ideas of local linear neuro-fuzzy models. The concept is presented, and the wide variety of existing learning schemes is summarized. The equivalence to Takagi-Sugeno fuzzy systems is analyzed, and the constraints for a proper interpretation in terms of fuzzy logic are outlined. Then the problem of learning is subdivided into two parts: (i) parameters of the local models and (ii) structural parameters of the rule premises. For the relatively simple part (i), the local and global estimation approaches are detailed and compared. For the much more complex part (ii), an overview of proposed learning schemes is given, and a specific algorithm based on incremental axis-orthogonal tree construction is discussed in detail. It is called local linear model tree (LOLIMOT) and will be used extensively throughout this book. Its basic features and settings are treated in this chapter. LOLIMOT allows training a local linear neuro-fuzzy model from data deterministically and without adjusting many fiddle parameters as it is usually the case for most alternative neural network approaches.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Strictly speaking, owing to the existence of an offset term, the LLMs are local affine not local linear. Nevertheless, “local linear models” is the standard terminology in the literature.
- 2.
In this context, “local” refers to the effect of the parameters on the model; there is no relation to the expression “nonlinear local” optimization, where “local” refers to the parameter search space.
- 3.
Note that equal data distribution is not always desirable; see Sect. 14.9.
- 4.
This step is necessary because all validity functions are changed slightly by the division as a consequence of the common normalization denominator in (13.4).
- 5.
Some slow-down effect can be observed because the loss function evaluations in (13.38) and Step 3 becomes more involved as the number of LLMs increases. However, for most applications, the computational demand is dominated by the parameter optimizations, and this slow-down effect can be neglected.
- 6.
At some point the number of training data samples will not suffice to estimate the parameters. Note that because of the regularization effect of the local estimation, this point is far beyond M = 150 where the nominal number of parameters is equal to the number of data samples.
- 7.
Both curves for σ n = 0 are, of course, identical since the training data is equal to the true process behavior.
References
Babuška, R.: Fuzzy Modeling and Identification. Ph.D. thesis, Dept. of Control Engineering, Delft University of Technology, Delft, The Netherlands (1996)
Babuška, R., Setnes, M., Kaymak, U., van Naute Lemke, H.R.: Simplification of fuzzy rule bases. In: European Congress on Intelligent Techniques and Soft Computing (EUFIT), pp. 1115–1119, Aachen, Germany (1996)
Babuška, R., Verbruggen, H.B.: Comparing different methods for premise identification in Sugeno-Takagi models. In: European Congress on Intelligent Techniques and Soft Computing (EUFIT), pp. 1188–1192, Aachen, Germany (1994)
Babuška, R., Verbruggen, H.B.: An overview of fuzzy modeling for control. Control Eng. Pract. 4(11), 1593–1606 (1996)
Babuška, R., Verbruggen, H.B.: Fuzzy set methods for local modelling and identification. In: Murray-Smith, R., Johansen, T.A. (eds.) Multiple Model Approaches to Modelling and Control, chapter 2, pp. 75–100. Taylor & Francis, London (1997)
Bernd, T., Kroll, A., Schwarz, H.: Approximation nichtlinearer dynamischer Prozesse mit optimierten funktionalen Fuzzy-Modellen. In: 7. Workshop “Fuzzy Control” des GMA-UA 1.4.2, Dortmund, Germany (1997)
Billings, S.A., Voon, W.S.F.: Piecewise linear identification of non-linear systems. Int. J. Control 46(1), 215–235 (1987)
Breiman, L., Stone, C.J., Friedman, J.H., Olshen, R.: Classification and Regression Trees. Chapman & Hall, New York (1984)
Carpenter, G., Grossberg, S.: The ART of adaptive pattern recognition by a self-organizing neural network. IEEE Comput. 21(3), 77–88 (1988)
Cleveland, W.S., Devlin, S.J., Grosse, E.: Regression by local fitting: methods, properties, and computational algorithms. J. Econ. 37, 87–114 (1996)
Ernst, S.: Hinging hyperplane trees for approximation and identification. In: IEEE Conference on Decision and Control (CDC), pp. 1261–1277, Tampa, USA (1998)
Fischer, T., Nelles, O.: Merging strategy for local model networks based on the Lolimot algorithm. In: International Conference on Artificial Neural Networks, pp. 153–160. Springer (2014)
Friedman, J.H.: Multivariate adaptive regression splines (with discussion). Ann. Stat. 19(1), 1–141 (1991)
Halme, A., Visala, A., Zhang, X.-C.: Process modelling using the functional state approach. In: Murray-Smith, R., Johansen, T.A. (eds.) Multiple Model Approaches to Modelling and Control, chapter 4, pp. 121–144. Taylor & Francis, London (1997)
Hathaway, R.J., Bezdek, J.C.: Switching regression models and fuzzy clustering. IEEE Trans. Fuzzy Syst. 1(3), 195–204 (1993)
Hilhorst, R.A.: Supervisory Control of Mode-Switch Processes. Ph.D. thesis, Electrical Engineering Department, University of Twente, Twente, The Netherlands (1992)
Hunt, K.J., Haas, R., Murray-Smith, R.: Extending the functional equivalence of radial basis functions networks and fuzzy inference systems. IEEE Trans. Neural Netw. 7(3), 776–781 (1996)
Isaksson, A.J., Ljung, L., Strömberg, J.-E.: On recursive construction of trees as models of dynamic systems. In: IEEE International Conference on Decision and Control, pp. 1686–1687, Brighton, UK (1991)
Ishibuchi, H., Nozaki, K., Yamamoto, N., Tanaka, H.: Construction of fuzzy classification systems with rectangular fuzzy rules using genetic algorithms. Fuzzy Sets Syst. 65, 237–253 (1994)
Jacobs, R., Jordan, M.I., Nowlan, S.J., Hinton, G.E.: Adaptive mixture of local experts. Neural Comput. 3, 79–87 (1991)
Jang, J.-S.R.: Neuro-Fuzzy Modeling: Architectures, Analyses, and Applications. Ph.D. thesis, EECS Department, Univ. of California at Berkeley, Berkeley, USA (1992)
Jang, J.-S.R.: ANFIS: adaptive-network-based fuzzy inference systems. IEEE Trans. Syst. Man Cybern. 23(3), 665–685 (1993)
Jang, J.-S.R., Sun, C.T., Mizutani, E.: Neuro-Fuzzy and Soft Computing: A Computational Approach to Learning and Machine Intelligence. Prentice Hall, Englewood Cliffs (1997)
Johansen, T.A.: Operating Regime Based Process Modeling and Identification. Ph.D. thesis, Dept. of Engineering Cybernetics, Norwegian Institute of Technology, Trondheim, Norway (1994)
Johansen, T.A.: Identification of non-linear system structure and parameters using regime decomposition. Automatica 31(2), 321–326 (1995)
Johansen, T.A., Murray-Smith, R.: The operating regime approach to nonlinear modelling and control. In: Murray-Smith, R., Johansen, T.A. (eds.) Multiple Model Approaches to Modelling and Control, chapter 1, pp. 3–72. Taylor & Francis, London (1997)
Jones, R.D., Coworkers: Nonlinear Adaptive Networks: A Little Theory, a Few Applications. Technical Report 91-273, Los Alamos National Lab., New Mexico (1991)
Jordan, M.I., Jacobs, R.A.: Hierarchical mixtures of experts and the EM algorithm. Neural Comput. 6(2), 181–214 (1994)
Kortmann, P.: Fuzzy-Modelle zur Systemidentifikation. Reihe 8: Mess-, Steuerungs- und Regelungstechnik, Nr. 647. VDI-Verlag, Düsseldorf (1997)
Kroll, A.: Fuzzy-Systeme zur Modellierung und Regelung komplexer technischer Systeme. Reihe 8: Mess-, Steuerungs- und Regelungstechnik, Nr. 612. VDI-Verlag, Düsseldorf (1997)
Ljung, L.: System Identification: Theory for the User, 2nd edn. Prentice Hall, Englewood Cliffs (1999)
Meila, M., Jordan, M.I.: Markov mixtures of experts. In: Murray-Smith, R., Johansen, T.A. (eds.) Multiple Model Approaches to Modelling and Control, chapter 5, pp. 145–166. Taylor & Francis, London (1997)
Murray-Smith, R.: A Local Model Network Approach to Nonlinear Modeling. Ph.D. thesis, University of Strathclyde, Strathclyde, UK (1994)
Murray-Smith, R., Johansen, T.A.: Local learning in local model networks. In: IEE International Conference on Artificial Neural Networks, pp. 40–46 (1995)
Murray-Smith, R., Johansen, T.A.: Local learning in local model networks. In: Murray-Smith, R., Johansen, T.A. (eds.) Multiple Model Approaches to Modelling and Control, chapter 7, pp. 185–210. Taylor & Francis, London (1997)
Nakamori, Y., Ryoke, M.: Identification of fuzzy prediction models through hyperellipsoidal clustering. IEEE Trans. Syst. Man Cybern 24(8), 1153–1173 (1994)
Narendra, K.S., Balakrishnan, J., Ciliz, M.K.: Adaptation and learning using multiple models, switching, and tuning. IEEE Control Syst. 15(3), 37–51 (1995)
Nelles, O.: LOLIMOT – Lokale, lineare Modelle zur Identifikation nichtlinearer, dynamischer Systeme. Automatisierungstechnik (at) 45(4), 163–174 (1997)
Nelles, O.: Nonlinear System Identification with Local Linear Neuro-Fuzzy Models. Ph.D. Thesis, TU Darmstadt, Automatisierungstechnik series. Shaker Verlag, Aachen, Germany (1999)
Nelles, O., Isermann, R.: Basis function networks for interpolation of local linear models. In: IEEE Conference on Decision and Control (CDC), pp. 470–475, Kobe, Japan (1996)
Omohundro, S.M.: Efficient algorithms with neural network behavior. J. Complex Syst. 1, 273–347 (1987)
Omohundro, S.M.: Bumptrees for efficient function, constraint and classification learning. In: Lippmann, R.P., Moody, J.E., Touretzky, D.S. (eds.) Advances in Neural Information Processing Systems, pp. 693–699. Morgan Kaufmann, San Francisco (1991)
Park, M.-K., Ji, S.-H., Kim, E.-T., Park, M.: Identification of Takagi-Sugeno fuzzy models via clustering and hough transform. In: Hellendoorn, H., Driankov, D. (eds.) Fuzzy Model Identification: Selected Approaches, chapter 3, pp. 91–161. Springer, Berlin (1997)
Pickhardt, R.: Adaptive Regelung nach einem Multi-Modell-Verfahren. Reihe 8: Mess-, Steuerungs- und Regelungstechnik, Nr. 499. VDI-Verlag, Düsseldorf (1995)
Pickhardt, R.: Adaptive Regelung auf der Basis eines Multi-Modells bei einer Transportregelstrecke für Schüttgüter. Automatisierungstechnik 3, 113–120 (1997)
Pottmann, M., Unbehauen, H., Seborg, D.E.: Application of a general multi-model approach for identification of highly nonlinear processes: a case study. Int. J. Control 57(1), 97–120 (1993)
Quinlan, J.: C4.5 Programs for Machine Learning. Morgan Kaufmann Publishers, San Francisico (1993)
Reed, R.: Pruning algorithms: a survey. IEEE Trans. Neural Netw. 4(5), 740–747 (1993)
Sanger, T.D.: A tree-structured adaptive network for function approximation in high-dimensional spaces. IEEE Trans. Neural Netw. 2(2), 285–293 (1991)
Sanger, T.D.: A tree-structured algorithm for reducing computation in networks with separable basis functions. Neural Comput. 3(1), 67–78 (1991)
Strokbro, K., Umberger, D.K., Hertz, J.A.: Exploiting neurons with localized receptive fields to learn chaos. J. Complex Syst. 4(3), 603–622 (1990)
Sugeno, M., Kang, G.T.: Structure identification of fuzzy model. Fuzzy Sets Syst. 28(1), 15–33 (1988)
Sugeno, M., Tanaka, K.: Successive identification of a fuzzy model and its application to prediction of a complex system. Fuzzy Sets Syst. 42, 315–334 (1991)
Takagi, T., Sugeno, M.: Fuzzy identification of systems and its application to modeling and control. IEEE Trans. Syst. Man Cybern. 15(1), 116–132 (1985)
Tanaka, M., Ye, J., Tanino, T.: Identification of nonlinear systems using fuzzy logic and genetic algorithms. In: IFAC Symposium on System Identification, pp. 301–306, Copenhagen, Denmark (1994)
Titterington, D., Smith, A.F.M., Markov, U.E.: Statistical Analysis of Finite Mixture Distributions. Wiley, Chichester (1985)
Wang, L.-X.: Adaptive fuzzy systems and control. design and stability analysis. Prentice Hall, Englewood Cliffs (1994)
Yoshinari, Y., Pedrycz, W., Hirota, K.: Construction of fuzzy models through clustering techniques. Fuzzy Sets Syst. 54, 157–165 (1993)
Zhang, X.-C., Visala, A., Halme, A., Linko, P.: Functional state modelling approach for bioprocesses: local models for aerobic yeast growth processes. J. Process Control 4, 127–134 (1994)
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2020 The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Nelles, O. (2020). Local Linear Neuro-Fuzzy Models: Fundamentals. In: Nonlinear System Identification. Springer, Cham. https://doi.org/10.1007/978-3-030-47439-3_13
Download citation
DOI: https://doi.org/10.1007/978-3-030-47439-3_13
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-47438-6
Online ISBN: 978-3-030-47439-3
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)