Showing papers by "Robert Babuska published in 1999"
TL;DR: It is shown that the fuzzy model is not only more accurate but as opposed to other black-box approaches (such as neural networks), it also provides some insight into the nonlinear relationship represented by the model.
320 citations
TL;DR: Recent work focusing on the use of Takagi–Sugeno fuzzy models in combination with MBPC is described, including a branch-and-bound method with iterative grid-size reduction and control based on a local linear model.
139 citations
01 Oct 1999
TL;DR: A novel approach to nonlinear classification is presented, in the training phase of the classifier, the training data is first clustered in an unsupervised way by fuzzy c-means or a similar algorithm, and a fuzzy relation between the clusters and the class identifiers is computed.
Abstract: A novel approach to nonlinear classification is presented, in the training phase of the classifier, the training data is first clustered in an unsupervised way by fuzzy c-means or a similar algorithm. The class labels are not used in this step. Then, a fuzzy relation between the clusters and the class identifiers is computed. This approach allows the number of prototypes to be independent of the number of actual classes. For the classification of unseen patterns, the membership degrees of the feature vector in the clusters are first computed by using the distance measure of the clustering algorithm. Then, the output fuzzy set is obtained by relational composition. This fuzzy set contains the membership degrees of the pattern in the given classes. A crisp decision is obtained by defuzzification, which gives either a single class or a "reject" decision, when a unique class cannot be selected based on the available information. The principle of the proposed method is demonstrated on an artificial data set and the applicability of the method is shown on the identification of live-stock from recorded sound sequences. The obtained results are compared with two other classifiers.
105 citations
01 Mar 1999
TL;DR: Tutorials - an overview of fuzzy modelling and model-based fuzzy control, R. Babuska fuzzy control - from heuristic PID to optimization-based nonlinear control, K.-E.
Abstract: Tutorials - an overview of fuzzy modelling and model-based fuzzy control, R. Babuska fuzzy control - from heuristic PID to optimization-based nonlinear control, K.-E. Arzen and M. Johansson surveys on advanced methodologies - neuro-fuzzy methods, D. Nauck complexity reduction methods for fuzzy systems design, M. Setnes fuzzy decisions for control systems, U. Kaymak fuzzy techniques in fault detection, isolation and diagnosis, P.M. Frank and T. Marcu applications - high level process control in the cement industry by fuzzy logic, J.-J. Ostergaard fuzzy logic for paper and pulp management, H. Hellendoorn implementing fuzzy control in the manufacture of washing powders, M. Setnes semi-mechanistic modelling and its application to biochemical processes, H.A.B. te Braake et al fuzzy control of a filling system in packaging machinery, C. Fantuzzi and F. Basissi fault detection and diagnosis of electro-mechanical drives and actuators, R. Isermann and D. Fuessel fuzzy logic for traffic management and control, H. Hellendoorn fuzzy modelling and control in avionics, G. Schram et al medical applications of fuzzy logic control, D.A. Linkens and M.F. Abbod linear design approach to a fuzzy controller, J. Jantzen.
63 citations
01 Jan 1999
TL;DR: Fuzzy Logic Applications in Mobile Robotics A. Ollero, et al., and Intelligent Data Analysis and Fuzzy Control H.-J.
Abstract: Preface. Part I: The Position and State of the Art of Fuzzy Systems. 1. Fuzzy Systems in Control Engineering H.B. Verbruggen, P.M. Bruijn. 2. Fuzzy Logic, Control Engineering and Artificial Intelligence D. Dubois, et al. 3. Fuzzy Control Versus Conventional Control K.-E. Arzen, et al. 4. Data-Driven Construction of Transparent Fuzzy Models R. Babuska, M. Setnes. Part II: Design and Analysis Issues. 5. Fuzzy Logic Normal Forms for Control Law Representation I. Perfilieva. 6. Stability Analysis of Fuzzy Control Loops A. Ollero, et al. 7. Performance Criteria: Classical and Fuzzy Design J.M. Sousa, et al. 8. Complexity Reduction Methods for Fuzzy Systems M. Setnes, et al. Part III: Application of Fuzzy Systems. 9. Intelligent Data Analysis and Fuzzy Control H.-J. Zimmermann, et al. 10. Fuzzy Control in Process Industry E.K. Juuso. 11. Fuzzy Logic Applications in Mobile Robotics A. Ollero, et al. 12. Enhancing Flight Control using Fuzzy Logic G. Schram, et al. References. Index.
51 citations
TL;DR: In this paper, two types of fuzzy logic controllers were developed for intermittent aeration control: a low-level fuzzy controller for DO control and a high-level controller for nitrogen removal.
40 citations
TL;DR: This article presents a modeling approach that is based on a combination of first-principle modeling of known relationships, with fuzzy modeling of the unknown parts of a process, and shows that the approach leads to an accurate prediction model, and allows for a qualitative interpretation of theunknown relationships learnt from data.
31 citations
01 Aug 1999
TL;DR: An overview of nonlinear predictive control based on neural and fuzzy models and several optimization approaches within the predictive controller based on these nonlinear model structures are reviewed, including iterative methods, operating-point and feedback linearization, and discrete search techniques.
Abstract: An overview of nonlinear predictive control based on neural and fuzzy models is given. The similarities and differences of these two modeling approaches are discussed, as well as their advantages and drawbacks. Several optimization approaches within the predictive controller based on these nonlinear model structures are reviewed, including iterative methods, operating-point and feedback linearization, and discrete search techniques. Some applications are reviewed and an example is given.
24 citations
01 Mar 1999
17 citations
01 Jan 1999
TL;DR: An algorithm for incorporating of a priori knowledge into data-driven identification for dynamic fuzzy models of the Takagi-Sugeno type by using input-output data and finding optimal parameter values by means of quadratic programming is presented.
Abstract: This paper presents an algorithm for incorporating of a priori knowledge into data-driven identification for dynamic fuzzy models of the Takagi-Sugeno type. Knowledge about the modeled process such as its stability minimal or maximal static gain, or the settling time of its step response can be translated into inequality constraints on the consequent parameters. By using input-output data, optimal parameter values are then found by means of quadratic programming. The proposed approach was successfully applied to the identification of a laboratory liquid level process.
14 citations
Journal Article•
01 Jan 1999
TL;DR: Fuzzy sets, the foundation of fuzzy control, were introduced thirty years ago as a way of expressing non-probabilistic uncertainties and found applications in database management, operations analysis, decision support systems, signal processing, data classifications, computer vision, etc.
Abstract: Fuzzy sets, the foundation of fuzzy control, were introduced thirty years ago, (Zadeh, 1965), as a way of expressing non-probabilistic uncertainties. Since then, fuzzy set theory has developed and found applications in database management, operations analysis, decision support systems, signal processing, data classifications, computer vision, etc. The application area that has attracted most attention is, however, control. In 1974, the first successful application of fuzzy logic to control was reported (Mamdani, 1974). Control of cement kilns was an early industrial application (Holmblad and Ostergaard, 1982). Since the first consumer product using fuzzy logic was marketed in 1987, the use of fuzzy control has increased substantially. A number of CAD environments for fuzzy control design have emerged together with VLSI hardware for fast execution. Fuzzy control is being applied industrially in an increasing number of cases, e.g., (Froese, 1993; Hellendoorn, 1993; Bonissone, 1994; Hirota, 1993; Terano et al., 1994).
01 Aug 1999
TL;DR: Three nonlinear predictive control algorithms were applied in simulation to a MIMO waste-water treatment process based on instantaneous linearization of the nonlinear prediction model and one is based on a branch-and-bound search technique.
Abstract: Three nonlinear predictive control algorithms were applied in simulation to a MIMO waste-water treatment process. Two algorithms are based on instantaneous linearization of the nonlinear prediction model and one is based on a branch-and-bound search technique. The prediction model employed is a fuzzy model of the Takagi-Sugeno type. The performance of the controllers is compared in terms of setpoint tracking, control effort and computational costs.
01 Jan 1999
01 Jan 1999
TL;DR: The objective of this chapter is to show that neural networks and fuzzy models can be incorporated in a semi-mechanistic modeling environment in a straightforward manner, and the main ideas, conclusions, and drawbacks will certainly hold for other application areas as well.
Abstract: The objective of this chapter is to show that neural networks and fuzzy models can be incorporated in a semi-mechanistic modeling environment in a straightforward manner. The procedure is described by the development of a semi-mechanistic model for a real biochemical process, the enzymatic conversion of Penicillin-G. Two different type of black-box model parts:(i) a fuzzy model, and (ii) a neural network, are compared. Finally, the semi-mechanistic model is used to create enough data to make a neural network model for the full process, which fits better in real-time control strategies. Although the results do not carry over directly to other engineering fields, the main ideas, conclusions, and drawbacks will certainly hold for other application areas as well.
01 Jan 1999
TL;DR: A time varying linear representaion is obtained which is used in LMBPC and Takagi-Sugeno (TS) fuzzy models are chosen, because the model structure as local linear models can be derived from the linear rule consequences in a direct way.
Abstract: MBPC is a nice technique to control multivariable systems while dealing with constraints and certain objective. Linear MBPC (LMBPC) is currently a settled theory and can be applied straightforward for linear processes. In this paper we deal with nonlinear systems, for which linear models that can be extracted. This way a time varying linear representaion is obtained which is used in LMBPC. Di erent schemes to obtain such local linear models are assessed in the light of the achieved performance of the predictive controller. Takagi-Sugeno (TS) fuzzy models are chosen, because the model structure as local linear models can be derived from the linear rule consequences in a direct way.